Issuu on Google+


The Political Economy of Productivity: The Case of Chile Cristóbal Aninat José Miguel Benavente Ignacio Briones Nicolás Eyzaguirre Patricio Navia Jocelyn Olivari

April 2010

Inter-American Development Bank Department of Research and Chief Economist

The Political Economy of Productivity: The Case of Chile

Cristóbal Aninat* José Miguel Benavente** Ignacio Briones*** Nicolás Eyzaguirre** Patricio Navia* Jocelyn Olivari**

* Universidad Diego Portales ** Universidad de Chile *** Universidad Adolfo Ibáñez

Inter-American Development Bank 2010

Cataloging-in-Publication data provided by the Inter-American Development Bank Felipe Herrera Library Aninat, Cristóbal. The political economy of productivity. The case of Chile / Cristóbal Aninat, José Miguel Benavente, Ignacio Briones, Nicolás Eyzaguirre, Patricio Navia, Jocelyn Olivari. p. cm. (IDB working paper series ; 105) Includes bibliographical references. 1. Chile—Economic policy. 2. Economics—Political aspects—Chile. 3. Productivity bargaining—Chile. I. Benavente, José Miguel. II. Briones, Ignacio. III. Eyzaguirre, Nicolás. IV. Navia, Patricio. V. Olivari, Jocelyn. VI. Inter-American Development Bank. Research Dept. VII. Title. VIII. Series. HB74.P65 A55 2010

© Inter-American Development Bank, 2010

Documents published in the IDB working paper series are of the highest academic and editorial quality. All have been peer reviewed by recognized experts in their field and professionally edited. The information and opinions presented in these publications are entirely those of the author(s), and no endorsement by the Inter-American Development Bank, its Board of Executive Directors, or the countries they represent is expressed or implied. This paper may be freely reproduced provided credit is given to the Inter-American Development Bank.

Abstract * This paper analyzes the political economy of productivity-related policymaking in Chile following a political transaction cost model (Spiller and Tommasi, 2003; Murillo et al., 2008). The main findings indicate that i) the Chilean policymaking process (PMP) was successful in the 1990s in implementing productivityenhancing policies, but as the country moved to a higher stage of development, the PMP grew less adept at generating the more complex set of policies needed to increase productivity at this stage; and ii) the Chilean PMP is less transparent than previously thought (Aninat et al., 2008), thus allowing political actors to favor private interests without being punished by the electorate. This has become apparent as the more sophisticated reforms needed at this stage of development require a deeper and more consolidated democracy. JEL Classifications: L52, O25, O40 Keywords: Economic policy, Institutional reforms, Productivity, Pensions, Education, Innovation, State modernization, Competitiveness, Chile


This paper was prepared as part of the Inter-American Development Bank project “Industrial Policy in Latin America and the Caribbean.�


1. Introduction Chile has experienced outstanding economic growth since its transition to democracy. From 1990 to 2005, its GDP expanded at an annual average of 5.5 percent. GDP per capita (PPP, current US$) increased from 4,800 to 12,250. The country entered the upper-middle-income bracket, significantly transforming its socioeconomic structure. This growth process has two distinct sub-periods. In 1990-97, 1 the country averaged a growth rate of over 7 percent. Afterwards, growth rates dropped to an annual average of slightly more than 4 percent. During the first sub-period, growth was driven by productivity gains. After 1997, however, the country experienced very low total factor productivity (TFP) gains. This paper seeks to identify and explain the political economy factors driving the decline in productivity growth rates. We analyze the political economy mechanisms responsible for the decline in productivity growth rates using a modular approach which combines different theoretical modules to study the distinct but interrelated dimensions of the policymaking process of productivity-enhancing policies. Rather than focusing on microeconomic TFP analysis, we analyze the types of policies and institutions needed to enhance productivity at three basic levels of development: factordriven, efficiency-driven, and innovation-driven economies. Policies that generate growth at the first stage are the basic requirements of institutions, infrastructure, macroeconomic stability, and human capital investments in health and education. At the efficiency stage, policies should be oriented toward quality higher education and training, goods and labor market efficiency, financial market sophistication, and technological readiness. On the economic frontier, growth is generated by business sophistication and the capacity of firms to innovate. Sustaining high levels of growth becomes more difficult as countries reach higher levels of development. During the period under study, Chile advanced from the basic needs stage to the need to increase efficiency and generate innovations in order to maintain high productivity growth rates. To analyze the factors driving the decision-making process for productivity-related policies, we use the political transaction-cost model put forward by Spiller and Tommasi (2003), Spiller, Stein and Tommasi (2003), and Murillo et al. (2008) that suggests that the quality of policies is determined by the characteristics of the policymaking processes (PMP), which in turn are determined by the characteristics of institutional arrangements and social actors intervening 1

Accelerated growth started in the mid-1980s. Here, we focus on democratic policymaking, hence starting in 1990.


in the relevant policy area. We expand this model to include the relevance of the characteristics of the specific policy issue being addressed. The analysis of institutions and the policymaking game in Chile builds on the work by Aninat et al. (2003, 2008). We deepen the analysis on the workings of institutions, particularly regarding the legislative process and the internal party system, including candidate selection, financing, and the election of party leaders. The paper’s main insight on the political economy of productivity comes from detailed cases studies of six productivity-related policy episodes. Policymaking processes that are successful at generating growth at a low stage of development may not be as successful at higher stages, when policies are more complex. Some policies require the electorate to master more sophisticated information and grasp its implications. For example, for the electorate it might be easier to understand how to improve the coverage of education than to grasp the complex set of variables that account for improvements in the quality of education. The Chilean PMP was successful at implementing first-generation and some second-generation reforms, but as the country developed it grew increasingly less adept at generating efficiency-enhancing policies necessary to maintain high productivity growth. Despite its low level of corruption and a transparent congress as the main policymaking arena, Chile’s PMP is, in our view, less transparent than previously thought. The opaque internal workings of political parties, together with the low competitiveness of the binominal electoral system, obscure the visibility of the moves of political actors. This allows politicians to favor private interests over the preferences of their constituencies without facing electoral consequences. This is compounded by the strongly ideological orientation of the Chilean electorate and policymakers. When Chile was implementing factor-driven policies characterized by high public visibility, the weight of potential information asymmetries was lower, allowing for good policymaking. The higher complexity of efficiency and innovation-enhancing policies has rendered policymaking less capable of generating good policy outcomes. We do not deny the previous success of the political system and policymaking process. In fact, the new order problems are only possible because Chile was successful. Yet, we caution about the PMP’s limitations for addressing current policy challenges and thus generating future productivity (and economic) growth.


Finally, we found that the capacity for representation of policy interests by major socioeconomic actors is influenced by their level of aggregation in relevant policy areas and by the ideological intensity and fragmentation of the actors and their historical public standing. Examples of this are the right-wing think tank Libertad y Desarrollo (LyD) and the labor union Confederación Unitaria de Trabajadores (CUT). LyD, a think tank financed by large corporations and wealthy individual donors, has an enormous influence on the policy process. Although widely associated with the most conservative right-wing party, the Independent Democratic Union (UDI), LyD does not have formal ties to the right-wing electorate or to the community of small and medium-sized business owners. But given its members’ role in the establishment of the market-based economic system during the military regime, it enjoys the support of the business-minded electorate, even when LyD defends the interests of large corporations and majority shareholders against the interests of SMEs and minority stock owners. Similarly, CUT has considerable political influence despite Chile’s low unionization rate and its small number of members. Given its role during the dictatorship, and the general popular perception that workers are too vulnerable under the current economic system, CUT enjoys workers’ support even when it defends the interests of well-paid CODELCO (state-owned copper company) workers, adversely affecting the interest of the general working class. Public opinion tends to associate the CUT with left-wing parties. The policy episodes discussed below present further evidence to illustrate these conclusions. In what follows, we describe the analytical frameworks used in the paper. Sections 3 and 4 present the productivity story of Chile in light of these frameworks. Before presenting our six case studies, we analyze productivityrelevant institutions and actors and their effects on the policymaking process with respect to productivity.

2. The Policymaking Process and Policy Outcomes Our political economy analytical framework is based on the political transaction-cost model proposed by Spiller and Tommasi (2003, 2007) and Murillo et al. (2008). Our analysis of productivity determinants is based on the framework proposed by Xavier Sala-i-Martín in his work for the World Economic Forum Competitiveness Index (2008). Spiller and Tommasi (2003) argue that policy outcomes (Y) are determined by the characteristics of the policymaking process by which the policy is produced (designed, approved, and implemented). The lower the


transaction costs faced by political and social actors in their intertemporal policy exchanges, the higher the expected cooperation and the higher the quality of policy outcomes. In this view, policymaking games are specific to each policy area and result from the interaction of the general policymaking game of the country (X) with the characteristics of each policy area (Z). Good policy outcomes are those in which policies are stable when confronted with political shocks, flexible with respect to socioeconomic shocks, well coordinated, and highly regarded by the public. When there are low transaction costs and long time horizons, actors can cooperate to find the best policy responses to social issues. The characteristics of the PMPs (X) are given by the interaction of history and political and economic institutions. PMPs with low transaction costs result when that interaction generates few actors that interact repeatedly over long time horizons, in transparent arenas (observable moves) with relatively low short-term payoffs, and good enforcement mechanisms. Institutions and factors relevant for the PMP are the legal system, the electoral system, political party regulations (laws that regulate the formation of parties, their internal procedures, candidate selection, and party and campaign financing) transparency, conflict of interests, and lobbying legislation, norms regulating the legislative process, the balance of powers among the branches of government, the independence of the judiciary, the independence of the central bank, banking system regulations and anti-monopoly regulations, among others. Spiller and Tommasi (2007) and Murillo et al. (2008) enrich the model, identifying the effects that specific characteristics of social actors and of each policy area (Z) have on the relevant policymaking game and ultimately on policy outcomes. They focus on the characteristics of social actors (socioeconomic interests), the conditions and mechanisms that allow them to influence the PMP, the arenas where they interact, and the policy domains in which they choose to intervene. The Z-characteristics include the number of major social actors, their level of aggregation, their collective action capabilities and resources, the way in which they articulate their demands, the breadth of their policy interests, their time horizons, and the fragmentation—both spatial and temporal—of policy domains, among others. We contribute to the analytical framework by identifying mechanisms though which social actors and additional characteristics of policy issues affect the workings of policymaking processes within each policy area (Z). Given that the policies needed at different levels of development have different characteristics, the effects of policymaking processes over policy


outcomes are not linear. PMPs’ effects over policy outcomes are specific to levels of development and development strategies. The characteristics of the PMPs for specific policy areas and the specific characteristics of policy issues (ž) within these areas matter for the quality of policy outcomes. For this reason, we suggest that the appropriate unit of analysis is the specific policy issue, not the general policy area. As stated by Spiller, Stein and Tommasi (2003) and Murillo et al. (2008), the specific institutions that regulate a particular policy area, interacting with the institutions of the country, affect the dynamics of the specific PMP of that area, the participating actors, and transparency of movements, among other variables. However, the transaction costs of policy exchanges vary across policy issues within policy areas. Different policy issues involve different transaction costs and provide for different levels of transparency before public opinion. The complexity of policy issues together with the relative position of available policy responses vis-à-vis the status quo significantly affects the transaction costs of cooperation on that specific social issue.

The level of complexity of the issues—both the problems and the

solutions—affects the visibility of the relevant actors’ behavior. In this context, the visibility of the moves by political actors is not determined only by the transparency of the negotiating arenas but also by the complexity of the issue for the electorate. The complexity of the problems (the degree of public understanding of the resulting negative effects of a particular social situation) determines whether the issue enters the public debate. Higher-complexity issues may not be understood by the media and the citizenry. Thus, despite appearing in newspapers and on the evening news, they may quickly fade away. When the issues do not draw the attention of the public, interested social and political actors can reach backroom agreements over policies arrived at privately without paying any electoral or political costs. Again, in oversimplified terms, the public will likely pay more attention to a school that was badly built or school textbooks that were defectively printed than to a more complex policy reform which results in teachers being inadequately rewarded for their performance. The complexity of solutions (public understanding of the available policy responses to the social issue at hand) determines whether political actors are accountable to their constituencies for their policy positions once an issue has taken center stage in the public debate. The complexity of public understanding of the effects of the available policy responses is affected by whether or not there is a technical agreement on the most appropriate policy response to a


particular social situation, where the appropriateness is related to the ability of such a policy to serve the interests of the whole community; and by the relative standing of the available policy responses with vested interests and the general ideological leanings of the population over the issue at hand. The existence of technical consensus on a policy issue helps in making the behavior of political actors on behalf of narrow interests apparent to the electorate. But when citizens hold strong ideological views on policy issues, it is easier for political actors to defend vested interests on ideological grounds. As a result, political actors can make their moves politically less visible. Obscure policy issues, however, can be made more visible by “policy champions” or by external shocks that change the relative costs of different policy stances by decision makers. As illustrated by some of the policy episodes discussed below, political and social actors may affect the degree of public understanding of policy issues (both problems and solutions), therefore increasing their visibility. For example, a corruption scandal in 2003 created a window of opportunity that led to the approval of a state reform long sought without success by the rightwing coalition. In this context, the social issues most likely to generate good policy outcomes are those in which both the problem and the solution are visible to the electorate and can be easily related to voters’ wellbeing. In the case of mining, for example, historical developments create a widely shared perception that copper is a natural engine of national growth and development. Thus, the issue can be more effectively framed before public opinion than a more obscure and complex issue of commodities market reform or pension reform, even if those latter reforms have a more measurable and direct effect on growth, development, and people’s wellbeing. Ideological intensity and fragmentation are related variables that affect transaction costs of the general PMP. History determines the general ideological leaning of the population, which affects the ideological intensity and fragmentation of the electorate over policy issues. In Chile, the history of the 20 years prior to the formation of the post-Pinochet political system—growing political polarization in the 1960s, Allende’s “Chilean road to socialism,” and the military dictatorship’s human rights abuses and economic reforms—generated a highly ideological and polarized electorate. At one extreme, some voters believe that the largely unregulated economic model inherited from the military government is best for growth and that regulations and social protection are detrimental to the economy. At the other extreme, other voters think that the “Chicago Boys” economic model generated a ruthless brand of capitalism that has left the


population utterly unprotected and vulnerable to the business cycle and corporate greed. There is a bimodal distribution along these two extremes in which most of the population is located away from the center. (As explained below, this historical trend is reinforced by the current electoral system.) This distribution is reinforced and preserved by the electoral system, which generates two-party coalitions located away from the median voter. Ideological representation of legislative initiatives and policy reforms plays a significant role in policymaking. The right-leaning electorate’s policy perspective tends to oppose the implementation of policies that imply regulating the economy and increasing social protection. Conversely, the leftist electorate holds the opposite view, that any increase in regulation and social protection is good for the population. Before public opinion, right-wing political views can be characterized as advocating a smaller state and left-wing views as advocating a larger state. Thus, even when a debate entails no significant change to the public sector’s share of GDP, public debates are framed in terms of one extreme advocating for a greater role for the state (estatistas) while the other extreme is caricatured as seeking to shrink the state even further (neoliberals). This polarized ideological distribution has obscured the public policy debate—especially in areas that by definition and nature are inevitably complex and even obscure—and facilitated the capture of the two main political coalitions by large corporations, on the one hand, and by organized labor on the other. Under free-market arguments, right-wing parties have been able to implement pro-business rather than pro-market policies, while the Concertación has tended to favor unions over general workers and the poor. This phenomenon of obscuring the policy debate on ideological grounds has decreased the visibility of social problems and policy solutions across the board. The likelihood of political cooperation over policy issues is also affected by the spatial position of available solutions vis-à-vis the status quo and actors’ preferences. The larger the win set, the higher the chances for cooperation. When a large number of actors are affected by the possible policy solution, high complexity of the distribution of benefits and costs of the reform increases the difficulty of reaching an agreement. In the education reform discussed below, there were several actors affected in very different ways; thus it was extremely difficult to design a reform package to compensate the actors who were negatively affected.


A related factor affecting the transaction costs of reforms is whether specific exchange currencies exist to compensate the negatively affected interests. For example in the state reform case study, the government coalition was negatively affected by the introduction of a meritocratic system for senior civil servants, as it lost the possibility to politically appoint them. The reform introduced simultaneously compensation in the form of public financing for political campaigns.

3. Productivity-Enhancing Policies and Stages of Development Since Solow (1957), most methodologies that measure aggregate productivity have concentrated on calculating Total Factor Productivity (TFP). In essence, TFP is a residual from an accounting process embracing several, perhaps too many, aspects of economic performance. The number obtained is not very useful for deciding what to do to enhance productivity. Recently, competitiveness literature has suggested several indicators which, because they are derived from a comparative analysis, provide a more accurate picture of a country’s current competitive situation. Probably the best known is the World Economic Forum’s (WEF) Competitiveness Index, based on 12 different pillars. It is produced for more than a hundred countries around the world. The WEF methodology suggests that economies should be seen as evolving agents that pass through different stages of development. As suggested by Sala-i-Martín et al. (2008), in the first stage, an economy is factor-driven and countries normally compete based on their factor endowments, primarily unskilled labor and natural resources. Then, depending primarily on wage levels, which reflect companies’ productivity, countries move into the efficiency-driven stage of development, when they must begin to develop more efficient production processes and increase product quality. At this point, competitiveness is increasingly driven by higher education, efficient goods markets, and sophisticated financial markets. As countries move into the innovation-driven stage, they can sustain higher wages and higher living standards only if their businesses are able to compete with new products. At this stage, companies must compete through innovation, producing new and different goods using the most sophisticated production processes. The concept of stages of development is integrated into the Global Competitiveness Index (GCI) by assigning greater weight to those pillars that are relatively more relevant for a country


given its particular stage of development. Then, the 12 pillars are organized into three subindexes, each critical to a particular stage of development. The basic requirements sub-index groups those pillars most critical for countries in the factor-driven stage (institutions, infrastructure, macroeconomic stability, health, and primary education). The efficiency enhancers sub-index includes those pillars critical for countries in the efficiency-driven stage (higher education and training, goods market efficiency, labor market efficiency, financial market sophistication, technological readiness and market size). Finally, the innovation and sophistication factors sub-index includes the pillars critical to countries in the innovation-driven stage. In the following sections, we focus on the types of policies implemented in the last 20 years in light of the WEF analytical framework. Then, we analyze the institutions, actors, and workings of the policymaking process for productivity-related policies. Finally, we discuss six policy episodes that represent different types of policies and characterize the different dynamics present in the PMP.

4. Productivity in Chile since 1990 During the golden years of economic growth (1984-1997), Chile’s impressive average annual growth rate of over 7 percent was mainly driven by productivity gains. In contrast, no productivity gains occurred during 1960-1985, and the country experienced very low TFP gains after 1997. As a result, GDP growth has declined as well. After 1997, the TFP decline seems to have responded to the absence of second- and third-generation socioeconomic reforms necessary to enhance productivity at higher levels of development. The basic economic reforms implemented correspond to what the WEF calls institutional reforms for a factor-driven economy. The first round of such reforms arose during the second half of the 1970s and established the basis for Chile’s current economic system. The country switched from serious macroeconomic imbalances and a high level of government intervention in the economy—including ISI (import substitution industrialization), basic price controls, financial market intervention, and state-owned companies in the financial markets and most economic activities—to a market-oriented economy. Major policies included the unilateral opening of the economy (setting a single tariff rate at 10 percent and a single exchange rate), price and wage liberalization, financial liberalization, privatization of public companies, and


fiscal spending moderation. This first phase corresponded to the so-called “normalization” period. The 1980s was marked by two major socioeconomic reforms with profound effects on long-term productivity: the replacement of the pay-as-you-go pension system by a private system based on individual capitalization accounts (AFP) and the transformation of the centralized public education system into a decentralized “voucher” system based on municipal and for-profit private schools. After the economic crisis of 1982, the country undertook a second wave of basic institutional reforms. The post-crisis period brought regulation of the financial market and an increase in the power vested in the Superintendency of Banks and Financial Institutions (SBIF). At the same time, the government divested most of its public utility companies (communication, energy), and an ambitious plan of privatization was launched. Finally, in 1989 the Central Bank of Chile became independent. The restoration of democracy in 1990 led to a third phase of economic reforms. The government complemented the unilateral trade liberalization of the 1980s with an extensive program of bilateral free trade agreements. Reforms also included the privatization of water utilities and the implementation of an ambitious and successful private infrastructure concession scheme for public works. A major judicial reform was also launched by the end of the 1990s, while reforms in capital markets (MK1, MK2) concerning IPOs and corporate governance were enacted. Additionally, the government strengthened its commitment to macroeconomic stability by launching a structural fiscal budget surplus rule. The Central Bank launched an inflationtargeting scheme, and the exchange rate regime moved gradually to a free float. Concertación governments expanded targeted social spending. Public policies coupled with strong economic growth led to sharp poverty reduction and improvement in health indicators. Education coverage expanded, but quality remains low by OECD standards. These policies consolidated and improved the system. The national policy debate therefore garnered major media attention and received significant public and political support from most relevant actors. They also generated considerable productivity gains. By the end of the decade, after implementing most structural reforms, it became apparent that an increasing number of policy reforms necessary to boost productivity were technically more complex. In terms of the WEF/Sala-i-Martín et al. (2008) categorization, the stage of development achieved by Chile required second- and third-generation reforms related to


efficiency gains and innovation. In order to become more competitive, an economy such as Chile’s requires sources of growth fueled by “inspiration” rather than “perspiration.” Chile is an outlier among Latin American countries. In 2008, it ranked 28th in competitiveness worldwide. According to Sala-i-Martin et al. (2008), this performance is mainly explained by its sound macroeconomic management, together with timely market liberalization and opening to trade, all taking place within the context of a transparent and predictable regulatory framework. Specifically, Chile has successfully laid most of the basic foundations for competitiveness, including strong macroeconomic fundamentals (14th), well-developed infrastructure (30th), efficient institutions (37th), and a good health-care system (31st in the health sub-pillar). Moreover, it displays efficient goods (26th) and labor (17th) markets. However there are some aspects where the country performance is inadequate. Both basic (105th) and higher (50th) education receive poor to middling grades, which bodes poorly for the country’s capacity for knowledge generation and innovation. On the innovation pillar, Chile ranks well below (45th) its country average.

5. Institutions, Actors, and the Policymaking Process 2 After summarizing Chile’s institutional arrangements and their effects on the policymaking process, we introduce a specific analysis of the characteristics of policies and the main social actors relevant for productivity related policymaking. A well designed system of checks and balances: Chile has a presidential system with a bicameral congress, a proportional electoral system with a district magnitude of two, and independent judiciary and other enforcement technologies. The system produces a PMP favorable to inter-temporal political transactions in most policy areas, as it generates a small number of relevant political actors who interact repeatedly with long time horizons, and there are good enforcement technologies overseeing the process. Checks and balances are designed to allow for slow but incremental change, preventing policy instability as a result of changes in the balance of political power. If these systems are well designed, however, they also allow for decisiveness when there are external shocks that change the relative costs of political issues. Checks and balances produce permanent and timely policy fine tuning in areas of relative consensus or in cases where compensation can be worked 2

This section draws upon Aninat et al. (2008), which provides a full discussion.


out, to have more status quo bias in more conflictive policy issues, and to produce relative gridlock on issues where political actors have strong opposing views. The current political institutional arrangements were established by the 1980 Constitution, which introduced provisions to strengthen the agenda-setting power of the president and to solve common pool problems on matters such as fiscal policy. The military government carried out a major transformation of the economic system from a centralized economy to a free market economy largely unregulated by 1982. The severe effects of the 1982 financial crisis led the government to introduce more regulations. The strong status quo bias of the policymaking process has allowed the minority right-wing political party coalition to block reform of the fundamentals of the economic system. The political process has been effective in protecting the main elements of the economic system while generating a stable but decisive policymaking process. It has allowed for an incremental improvement of policies, institutions, and rules and regulations. It has allowed for stability with respect to political shocks and for decisiveness with respect to socio-economic shocks. 5.1 Institutional Determinants of the Policymaking Process A unitary state with a centralized government: There is no fiscal decentralization in Chile. During the twentieth century, Chile’s bureaucracy developed a reputation for low corruption, but was also marked by strong centralization, an emphasis on procedures rather than on outputs, poor coordination mechanisms, and no participation of civil society or market mechanisms in the provision of public services, making it rigid, relatively fragmented and not very efficient. Economic transformations carried out by the military dictatorship in the 1970s started an incremental process of reform that led to significant reforms since the 1990s during the Aylwin (1990-94), Frei (1994-00), Lagos (2000-06), and Bachelet (2006-10) administrations, all members of the center-left Concertación coalition. The military implemented structural economic reforms and reduced the size of the state by privatizing public companies and the social security system, deregulating markets, and improving the tax system. These reforms, however, were carried out in an ideological environment contrary to empowering the state, which led to under-investing in institutional capacity and modernization of the delivery of public services.


After the end of the dictatorship, as both the nascent democracy and the new economic institutions were consolidated, gradual reform of the civil service continued to unfold. The most important legislation addressed simplifying administrative procedures, updating the level of remuneration of high public officials, reducing the amount of undisclosed funds, professionalizing the civil service in order to improve and develop long-term careers in the civil service regardless of political changes, and addressing access to information, transparency, lobbying and conflict of interest issues. Despite the effort to improve the capabilities of the state, the bureaucracy is still too rigid, fragmented, and procedure oriented. The state has poor coordination arrangements at the central government level. The office of the president has a small staff, politically appointed, and there are no permanent structures in charge of coordination of inter-sectoral programs. This fragmentation of the policy space is compounded by coalition governments. The electoral system leads to the formation of coalition governments in which the ministries are managed by cabinet members of different political parties. Strong presidential system: Chile has one of the strongest presidential systems in Latin America. The Executive has exclusive legislative initiative in several policy areas, a highly hierarchical control of the budget process, and an array of “urgency” and veto options, which make it a de facto agenda-setter. Yet, capable lawmakers and a long legislative process with supermajorities for most economic policies impose on the Executive the need to negotiate its legislative agenda. Presidents are elected for four-year terms and banned from running for immediate re-election. The nomination process has varied since 1990, but the two main coalitions have presented candidates for every election and have dominated the elections. All four presidential elections have been won by Concertación candidates. The president’s constitutionally mandated legislative and non-legislative powers are substantial. Its non-legislative powers include broad authority to nominate, appoint, and dismiss government officials. With Senate approval, the president can nominate Supreme Court justices, members to the Central Bank’s governing board, three of the ten members of the Constitutional Tribunal, and the Comptroller General. This procedure ensures a politically balanced nomination to key institutional posts. The president appoints all cabinet ministers, regional and provincial governors, ambassadors and heads of government agencies and state companies. Given that the electoral system fosters coalition governments, cabinet formation might be negatively affected by reducing the cabinet’s cohesion. However, the highly hierarchical structure of the cabinet,


combined with the president’s authority to appoint and dismiss ministers, and the prominent role of the finance minister in determining the budget of each ministry, has produced relatively efficient cabinets. The degree of support for the president in Congress is relatively high and constant under the current electoral system. The binomial electoral system leads to rather similar congressional representation for each coalition in Congress. In this context, support for the Executive in Congress oscillates around 50 percent in all legislatures. A bicameral Congress: Chile has a 38-member Senate and a 120-member Chamber of Deputies. Both chambers use the binominal electoral system, without term limits. Deputies are elected from 60 two-member districts for four-year terms, and senators are elected from 19 twomember districts for eight-year terms with a staggered renewal every four years. Two main institutions shape the workings of Congress: the executive boards (mesas) of each chamber and the chambers’ committees. Formal rules determine that committee assignments do not depend on seniority but on the mesas’ nomination and ratification by the respective chamber, and they proportionally reflect the partisan composition of each chamber. In actuality, however, the mesas consider the parties’ requests for committee assignments. Independent and politically insulated enforcement mechanisms: The main enforcement technologies overseeing the political system are the Judiciary, the Constitutional Tribunal and the Comptroller General (CGR). The Constitution establishes their independence from other branches of the government and their institutional design accomplishes that goal. All three institutions have nomination procedures that assure their political insulation and independence from other branches of the state and avoid partisan biases in their composition. The 21 Supreme Court justices are nominated by the president from a five-person list proposed by the Court, and have to be approved by two-thirds of the Senate. The Constitutional Tribunal is composed of 10 members—three chosen by the president, two by the Senate, two by the Chamber of Deputies with Senate approval, and three by the Supreme Court—for staggered nine-year terms. Three members are renewed every three years.

As a result of the 2005

constitutional reforms, the Tribunal now has more powers and responsibilities. Constitutional challenges can be brought to the Tribunal by the executive, legislative and judicial branches, but individuals can also challenge the constitutionality of some government actions. The rulings of the Tribunal are not subject to appeals, though it can rectify previous rulings if it independently decides to do so. The CGR is also autonomous and independent. The Comptroller General is


appointed by the president with Senate approval, and cannot be removed until he/she is 75 years old, unless impeached by the Senate based on a constitutional accusation by the Chamber of Deputies. Even though all three enforcement technologies are nominated by other powers of the state, a correct combination of checks and balances assures their political independence. In all cases, those who aspire to hold these offices cannot cater to the interests of one branch of the state alone and, once nominated, their non-removability insulates them politically. The electoral system, party legislation, and coalition politics: Chile has two different electoral systems for legislative and presidential elections, which give different incentives to political actors and cross-pressure them in their electoral strategies. Parliamentary elections occur under a proportional representation system with a district magnitude of two in all districts (the binominal system). It uses open lists, with two candidates per list, and the d’Hondt seat allocation formula. It allows for nationally binding coalition formation. Each coalition receiving the two highest vote shares wins one of the two seats per district unless the list receiving the most votes doubles the votes of its second-place rival, in which case it receives both seats. By introducing high barriers to entry, the electoral system has a two-fold effect on the party system. On the one hand, it reduces the number of relevant actors by encouraging parties to coalesce. On the other, it reduces electoral competition and strengthens the national leadership of parties. Since 1990, there have been six parties with congressional representation, organized into two national coalitions, with center and left-wing parties that opposed Pinochet in one coalition, the Concertación, and right-wing parties that supported him in the other, the Alianza. The Concertación and the Alianza have remained as the only two coalitions with congressional representation and have become the longest-standing coalitions in Chile’s long republican history. The ruling center-left Concertación comprises four parties (Christian Democratic PDC, Social-Democratic PPD, Socialist PS and Social Democratic PRSD). The right-of-center Alianza opposition is made up of two parties (the conservative Independent Democratic Union, UDI, and the more liberal National Renewal, RN). In addition, the Communist Party (PC) and Humanist Party (PH) also have local elected officials but no parliamentary representation. Though an overwhelming majority of legislators come from the Concertación and Alianza, independents have occasionally won seats as well.


The existence of coalitions provides political parties with several incentives and restrictions. Since only one-third of the vote is needed to secure half of the seats in each district, it drives parties’ platforms away from the median voter. This polarizing effect of the electoral system perpetuates the ideological divide of the electorate inherited from the 1988 Plebiscite. The mediating effects of the presidential electoral system reduce its centrifugal incentives. Also, the potentially high price a party would pay if it unilaterally left its coalition enforces high intracoalition discipline. This has helped maintain unity in both coalitions, since they include parties with different platforms on several issues and political leaders with high personal rivalries. Despite publicized intra-coalition bickering, both coalitions have remained united by the pressure of their legislators in their respective party’s leadership. Members of Congress know that their chances for re-election would be jeopardized should their coalitions break up. The binomial system also encourages continuous intra-coalition party negotiations to decide which candidates will be included in the coalition’s lists in every district, strengthening, to a certain extent, the national leadership of parties. The lack of regulations on candidate selection procedures has led to back-room negotiations, as opposed to primaries, increasing the opacity of parties’ internal workings. Each party has to negotiate with its partners the districts in which it presents candidates while responding to both national and local considerations. On the one hand, parties have to negotiate their share of the coalition’s candidate list, and on the other, they have to try to get into districts where the party has candidates with strong local support. In legislative elections, the binomial system normally gives each of the two main coalitions a secure seat in most districts. This in turn allows coalitions to present candidates responding to strategic national interests instead of voters’ preferences. This has a negative effect on the transparency of party policy preference formation. As parties do not need to cater to median voter preferences, in issues with low visibility they can represent private interests without risking electoral failure. Moreover, as candidates need to obtain the party nomination to run in the coalition’s list, parties can impose discipline on their behavior. As congressmen have long legislative careers, given no term-limit provisions and high entry barriers for challenges outside the two main coalitions, there is low leadership turnover. That in turn contributes to the perpetuation of the ideological orientations of parties. Seventy-five percent of members of Congress are re-nominated and about 60 percent are re-elected (Carey, 2002). The binomial system has three distinct effects: it reduces the number of relevant actors to a few parties


organized into two stable and disciplined coalitions, it reduces electoral competition, allowing for powerful and stable party leadership whose power is strengthened by the informal rules that determine the Congress’ authorities, and it provides high congressional support for the Executive. Presidential elections have a majoritarian electoral system with run-off provisions. This system induces the nomination of moderate candidates who seek the median voter position. Since only two large coalitions having congressional representation, this system induces intracoalition negotiations to nominate one candidate for each coalition, which in turn reinforces coalition cohesion and discipline. Similar intra-coalition horse trading has taken place when selecting the presidential candidate. In 1993 and 1999, Concertación held national primaries to select its presidential candidate. Yet in 2005 and 2009, presidential candidate selection took place without national primaries. Though polls were widely used by parties to bargain in favor of their candidates, the fact that those polls were taken in the absence of a formal campaign links the results to an exante popularity of a candidate rather than to an intention by voters. The combined effect of the binominal and presidential electoral systems on politicians, parties, and coalitions is a set of incentives which cross-pressure them and multiply the opportunities for intra-coalition leadership negotiation. The legislation for political parties established by the Pinochet regime has remained largely unchanged. It allows political parties to function in opacity, particularly regarding selection mechanisms and financing. Despite the existence of some legal transparency and procedural requirements for the selection of party officials, in practice party officials and candidates for public office are nominated behind closed doors. This, together with reduced militant base and low electoral competition, facilitates undue influence by special interests. 5.2 The Legislative Process: An Agenda-setting President vs. a Gate-keeping Congress Given the substantial agenda-setting powers of the president and a carefully designed balance of powers between the president and Congress, the president dictates the legislative agenda and Congress decides final approval of legislative initiatives. The president cannot advance his/her legislative agenda without the votes of a qualified majority for most major legislative reforms.


Contrary to other Latin American countries, there are no constitutional loopholes that allow the president to bypass Congress to approve legislation (Aninat, 2007). The constitutionally mandated agenda-setting powers enjoyed by the president are substantial. The Executive has sole legislative initiative over legislation concerning the political and administrative divisions of the state, its financial administration, the budget process, and the sale of state assets. Additionally, the Executive has sole initiative in areas such as taxation, labor regulation, social security, and the Armed Forces. The Executive has sole authority to initiate legislation that requires budget increases or allocation of new funds, which gives it exclusive legislative initiative over most economic policy areas. The Executive also controls the flow of legislation through the use of “urgencies”—a constitutional mechanism designed to give the Executive the power to force a rapid legislature vote on an initiative. The Ley Orgánica Constitucional del Congreso Nacional (#18,918) provides for three different types of urgencies—30, 10, and 3 days, respectively— that can be introduced or withdrawn at will by the Executive at any moment during the legislative process. Urgencies are used strategically by the government to test the potential legislative support for its initiatives and to control the order of the day in legislative committees and floor consideration of bills. Although it is weak when compared to the U.S. Congress, the Chilean legislature is unusually professional and technically competent by Latin American standards. The lack of term limits and high entry barriers for challengers allow legislators to develop long legislative careers. The legislative process consists of three “constitutional steps” (trámites constitucionales). The first step begins when a legislative initiative is submitted by the president to a chamber of her choice. Individual legislators and groups of legislators can also submit legislative initiatives, but they are rarely considered by Congress. The respective chamber assigns the initiative to one of its standing committees or to a specially assigned joint committee, depending on the nature of the initiative. The committee then discusses the bill “in general” and can modify the initiative at will. Regardless of the committee’s vote on the initiative, the chamber votes on the amended initiative presented to the floor by the committee. If amendments are offered during the general discussion, this triggers a second reading of the bill, called discussion in particular. At this stage, the bill is returned to the committee, and all amendments are discussed. Typically, the chamber will agree to extend the deadline for presenting amendments. Once this period has passed, the committee discusses and votes on the new amendments. The bill, along with any adopted


amendments, then returns to the floor. Any legislator can introduce amendments in advance of the second reading, and in the Senate, committee members must put each proposed amendment to a publicly reported vote. Once the first legislative step is completed, the initiative moves to the other chamber for the second legislative step. A similar process ensues. But since an initiative that cleared the first hurdle is closer to final passage, the second chamber often votes first on the intention to legislate (idea de legislar) the project as a package (en general) and in the end, the truly important vote is on the specific articles of the legislation after it has passed the respective committees. At this stage, if both chambers have approved the same version of the bill, it is sent to the president for his signature or veto. However, if the chambers have approved distinct versions of the bill (e.g., if the revising chamber has sustained any amendments) the bill is referred to a conference committee made up of an equal number of members from both chambers and chaired by a member of the Senate. For bills referred to a single committee, the Senate conferees are typically the members of the relevant committee. Deputies’ committees tend to be larger (13 members vs. 5 for the Senate), so not all of the deputies from the relevant committee are involved in the conference committee. The bill put forth by the conference committee is put to an up or down vote in each chamber, without the possibility of amendment. If and when both chambers have approved the same version of the bill, it goes to the president. She can sign the bill, in which case it often goes to the Constitutional Tribunal for approval of its consistency with existing constitutional provisions, if 15 members of Congress or the president so request or if provisions about the constitutionality of the objected part of the norm have been raised during the legislative process. Objections at this stage tend to be genuinely related to constitutional issues, so the Constitutional Tribunal does not act as a third legislative chamber. If the president does not wish to sign the bill, he can veto it in whole or in part and send it back to Congress. Congress must first vote (up or down, no amendments) on the “observations.” Only if these are rejected, Congress proceeds to vote on veto override, which for most laws requires the approval of three-fifths of the members present in each chamber. If the veto is overridden, or if “observations” are accepted, the bill is promulgated. If a successful partial presidential veto applies, then the remainder of the bill is promulgated. Constitutionally mandated supermajority thresholds for some special legislation contribute to offsetting the Executive’s substantial agenda-setting power. They range from an absolute


majority of the total membership (as opposed to a majority of those present) up to two-thirds of the members. Those provisions permit a minority in one chamber to block legislation. With the exception of the weakest threshold (a majority of total membership), the Concertación has always faced the need to negotiate changes to legislation that require supermajority approval with the conservative opposition in at least one legislative chamber. These high thresholds, together with the repeated interaction of the president with Congress for long time horizons, have led to tit-for-tat equilibrium legislative strategy. “Consensus politics” have emerged as the president negotiates with the opposition on bills requiring qualified majorities and most other legislative initiatives. 5.4 Major Actors in the Policymaking of Productivity In addition to the business and labor actors emphasized by Murillo et al. (2008), we identify other categories that can be as influential, such as think tanks and non-governmental organizations (NGOs). For each category we identify their preferences, the channels through which they exert political influence, and their influence on the policy outcomes of the case studies covered in the paper. Socioeconomic actors participate in the PMP in Chile at a high level of aggregation, either through business organizations, labor unions, or influential think tanks and NGOs. Business Organizations: The corporate sector is well organized in multilayered business federations that represent all areas of economic activity. They have high media standing; they interact repeatedly with the Executive and political parties in negotiations that affect their interests. Moreover, the lack of transparent and well-designed campaign financing mechanisms allows interest groups to exert unobservable influence on parties and think tanks’ legislative agendas. Other social actors include business associations and economic conglomerates. Since the mid twentieth century, a number of business associations have emerged to advocate for their particularistic interests. The Sociedad Nacional de Agricultura (SNA) brings together most large and mid-size agro-industrial entrepreneurs. The Cámara Chilena de la Construcción (CChC) brings together the construction and infrastructure sector. Altogether, six such business associations form the Confederación de la Producción y el Comercio (CPC), the association that defends the interests of the entrepreneurial class. After having actively participated in shaping


government policies in the period between 1940 and1960, as a result of governments that promoted import-substituting industrialization, these business sector organizations opposed the socialist Allende government (1970-73) and then enthusiastically supported the Pinochet dictatorship—but also opposed some liberalizing and market-friendly policies adopted under military rule. Then, these organizations slowly withdrew from an open political role when democracy was restored. In part because the policies championed by Concertación governments had produced positive results and also because it was in the best interest of their organizations to develop and maintain good relations with the government—regardless of the political color of the party in power—the business organizations made their own transition away from authoritarian rule. Most recently, the CPC has elected a former Allende supporter-turned businessman as its leader in 2008. The “size” and “formality” variables in Murillo et al. (2008) explain why the business sector is highly aggregated through very organized and powerful associations. As expected, SMEs are the main absent socioeconomic actors. Small firms across sectors probably hold different preferences and thus are organized, when they are, in different associations. The shorter-run approach of these associations makes coordination of collective action difficult. Hence, they tend to be underrepresented in the negotiating arenas around policy issues that affect them. As remarked in Murillo et al. (2008), large companies can be powerful enough to define policy preferences and policy influence. The pro-business rather than pro-market preferences of these organizations find a space in the negotiating arenas of the PMP mainly through a set of pressure mechanisms, which are likely to rely on a combination of economic and political tools. Their influence can be exerted through informal networks with political parties, the threat of disinvestment (and its impact in national economic variables), and/or financial inducements for electoral campaigns or personal enrichment. Informal networks between business associations and political parties can be of special importance for a small country with an unequal income distribution and where economic power is highly concentrated. A survey of managers of big companies revealed that 50 percent of them graduated from five schools, all of them private schools serving high-income families. Since only 7 percent of children are enrolled in such schools, business power is highly concentrated among the relative few that belong to the higher income group. According to Espinoza (2008),


60 percent of Chile’s political elite (particularly deputies) attended private schools, thus facilitating the development of a social network among businesspeople and political actors who grew up in the same milieu. Even though the document states that school networks are weaker than those developed through higher education, there are other sources of informal networking, or “secondary circles of sociability” that can be used as channels to exert influence. Labor unions: Historical episodes added to our specific economic conditions have led to stronger, more organized public sector federations. Teachers, public health sector employees, and public employees in general are well organized and play an important role in the electoral base of the Concertación. Some of these labor unions are: CUT (National Association of Unions), ANEF (National Federation of Public Sector Workers), the Teachers Union, the Health Workers Union, the Municipal Employees Union and the state-owned copper giant CODELCO workers union. The preferences of these socioeconomic players are related mainly to the maximization of job stability and salary increases based on seniority rather than performance. Concertación parties—most notably PDC and PS—have strong ties to labor unions. Public sector unions are among the strongest in Chile, including the powerful CODELCO union. Most unions are controlled by PS and PDC militants. The PC, to the left of the Concertación, also has a strong presence in the labor unions. In fact, PC militants led the CUT for several years in the 1990s. The CUT’s current president is a member of the Socialist Party, but the strength of the PC in the union movement often leads socialist militants to take positions to the left of the Concertación and even to the left of the PS within the Concertación. Other powerful labor unions include the Teachers Union (Colegio de Profesores) where the PC has commanding support, and the Health Professionals and Health Workers Unions (CONFUSAM), where the PS commands strongest support. The influence of unions within the party apparatus is far larger than their membership numbers would seem to indicate. Unions can mobilize organized voters and become crucial in helping man campaigns and providing avenues to disseminate information to politically sophisticated unionized workers. This is mainly due to their ideological features. Public workers (from the education, mining, health and state sector, among others) lean toward leftist ideology, so they are more prone to support organized leftist associations that defend, or seem to defend, public workers’ rights from a left-wing point of view. Thus, even though they are not formal members of these associations, they feel represented by them. This situation put these


socioeconomic players in a privileged arena for exerting influence over specific actors of the PMP (mainly left-wing political parties), especially when the problems and solutions in a specific policy area are less visible and more complex. The limited understanding by public workers of complex problems and solutions, added to their left-wing ideological leaning, gives enough space to unions to act under opacity, promoting actions related to vested interests at the expense of more technical solutions. Given the strong influence of unions over the citizens they represent, whether formally or informally, and the power they have to threaten political stability through demonstrations, political parties— especially the PS and PDC—find it difficult to adopt policy positions that will alienate their strong union base of support. Hence, political parties are the main channel through which these socioeconomic players influence the policymaking process. Think tanks and NGOs: Think tanks and NGOs play an active role in the policymaking process, both technical and political. Think tanks are independent and privately financed, but most are close to one or more political parties. There is a two-way relationship between the parties and think tanks. Think tanks provide legislative assistance in exchange for advancing their ideological agendas. The lack of congressional staffers to support the legislative work of legislators reinforces this relationship. Other social organizations, such as NGOs (Paz Ciudadana, the Chilean Chapter of Transparency International) and the churches (traditionally Catholic but increasingly others) play an important role in the aggregation and representation of social preferences in the PMP. Political parties are highly disciplined and organized, but they mostly lack technical expertise. Parties are much more concerned with organizing militants to run and win elections than with preparing technocrats who can design and analyze public policies. Thus, think tanks have emerged as the policy arm of the two coalitions. Given that the executive controls the legislative agenda, opposition parties can only react to government-sponsored legislation. Because bargaining is often necessary for the government to achieve majority and supermajority support in Congress to advance its legislative initiatives, opposition think tanks advise opposition legislators on which issues they should compromise on and on which items they should stick to their initial positions. In fact, it is often the case that government ministers negotiate directly with representatives of policy experts from opposition think tanks who also formally serve as legislative assistants for opposition senators and deputies.


In the Alianza, the Instituto de Libertad y Desarrollo (LyD) has evolved into the think tank that provides the UDI with policy recommendations and advice on how to vote on the government’s proposed legislation. For RN, Instituto Libertad (IL, a different think tank) serves the same purpose. Because it is better funded and staffed, and because it has had a more constant leadership, LyD has emerged as the most influential think tank within the Alianza. Also, because the UDI is a much more disciplined and top-down party, LyD can better influence the positions of UDI legislators than IL can the position of RN congresspersons. As non-profit corporations, think tanks can receive tax-deductible private funding from companies and interested parties but are not required to release information on the sources of funding they receive. There is plenty of anecdotal evidence—widely reported in newspapers and talked about among Concertación and Alianza legislators—that LyD receives funding from companies and entrepreneurs associated with the most conservative positions within the Catholic Church and with close ties to state companies privatized during the Pinochet dictatorship. LyD is the think tank most closely aligned with the neoliberal economic policies promoted by the Pinochet dictatorship. Former Finance Minister under Pinochet, Hernán Büchi serves on the Council of Advisors. Buchi’s former chief of staff in the Ministry of Finance, Cristián Larroulet, is the executive director of LyD. Former Chief of Cabinet of the Pinochet regime, Carlos Cáceres, presides over the Council of Advisors. Instituto Libertad, on the other hand, consists primarily of technocrats associated with RN. As a nonprofit corporation, IL gets funding from private sources as well. Its positions are regularly more liberal than those of LyD, but IL has less influence over RN legislators, produces fewer policy papers, and announces fewer policy positions than LyD. IL is also involved with helping organize campaigns for RN. In UDI, that role is mostly played by the Fundación Jaime Guzmán, a non-profit that occasionally also defends specific policy positions. In addition, the influential Centro de Estudios Públicos (CEP) serves as supplier of policy positions for right-wing legislators. The policy positions announced and defended by the independent CEP occasionally serve as a signaling device for right-wing legislators. CEP was originally created by the Matte family—one of the wealthiest family conglomerates in Chile, with financial interests in forestry, electricity, and almost every other sector of the Chilean economy. CEP is funded by contributions from private donations—mostly private companies— and its non-partisan ideology champions the defense of market-friendly policies and liberal


capitalism. Its influential Estudios Públicos policy journal, its occasional papers, and more indepth studies on state modernization, campaign finance, and economic policies have directly influenced debates. Occasionally, Concertación governments have sought to negotiate policy agreements with CEP in order to obtain legitimacy before presenting its legislative initiatives to Congress. Thus, the Concertación has used CEP to bypass the think tanks more directly associated with the Alianza. Think-tanks are also important for the Concertación. But since the center-left coalition has continuously occupied the executive branch in Chile since the restoration of democracy, technocrats are often lured into government positions from Concertación-friendly think tanks. For example, after Michelle Bachelet was elected president, she appointed three ministers— including the finance minister—from among members of the liberal-leaning and Concertaciónfriendly Expansiva think tank. Similarly, in 1990, President Patricio Aylwin recruited many of his technocrats—including his finance minister, Alejandro Foxley—from the Cieplan think tank. Thus, Concertación think tanks have failed to remain as permanent actors in the political arena. They form and, when successful, they are rapidly recruited into the executive. Because the executive effectively controls the legislative agenda and drafts most relevant legislative initiatives, Concertación think-tank technocrats have every incentive to join the government. As a result, Concertación think tanks are much weaker and more poorly staffed than Alianza think tanks. This adversely affects Concertación legislators, who have little technical support vis-à-vis the Concertación government when discussing policy reforms. Except for those legislators who develop a personal interest in specific policy issues, Concertación legislators can count on less technical support and expert advice when exercising their legislative role. 5.5 The Policymaking of Productivity in Chile The political institutions established by the military regime in the Constitution of 1980 were designed to generate a political system with a strong status quo bias in order to protect—under democratic politics—the dictatorship’s economic and institutional legacy. The ensuing PMP proved to be conducive to good and responsible policymaking, generating the largest economic expansion in Chilean history together with improvement in most social indicators (Aninat el al., 2008). The constitutional reform of 2005 eliminated the “undemocratic enclaves” but did not alter the incentive structure established by the original Constitution.


The complex system of checks and balances established by the Constitution imposes a structure of incentives and restrictions on actors’ behavior. It has two main effects on the policymaking process for productivity: a PMP generally conducive to inter-temporal cooperation and, within this cooperative framework, a strong status quo bias, since most policy reforms need the support of both government and opposition coalitions. As analyzed by Aninat et al. (2008), institutions generate a system with few actors, interacting repeatedly, over long time horizons, in mostly transparent negotiating arenas, and under the oversight of politically insulated enforcement mechanisms. Concentration of the agenda-setting power in the president facilitates public accountability, therefore inducing the president to internalize the general national interest. An essential aspect of the Chilean PMP is the centralization of agenda-setting power in the executive and the gatekeeping role played by the opposition in Congress (Aninat and Londregan, 2006). This balance of powers explains two central features of the PMP. It induces the president to internalize the country’s general interest and to propose a consensual policy agenda. The Constitution assigns most of the legislative initiative to the president in order to avoid common pool problemsp—and “Christmas tree” bills—at the same time that it empowers Congress to authorize the financing of this agenda. The clear responsibility of the executive for the socioeconomic management of the country increases political accountability. The balance of partisan power in Congress, on the other hand, contributes to repeated negotiationn between the executive and the opposition, leading to more consensual politics— what has been called “democracia de los acuerdos.” The Constitution establishes a delicate balance between the congressional representation of each coalition and the supermajorities required to approve all major socioeconomic reforms. The electoral system induces the formation of two large, disciplined party coalitions which obtain between 40 percent and 60 percent of the seats. The most relevant supermajority threshold for socioeconomic reforms is 4/7 (57.14 percent) of the members of each chamber. This implies that the government needs the consent of the opposition to pass major reforms. As explained above, given the repeated nature of their interaction, if the government imposes its preferred policies on areas that require simple majority it risks a “tit-for-tat” response of the opposition on qualified majority legislation. The combined institutional arrangement has helped consolidate a political agreement on responsible, non-populist macroeconomic management and the general socioeconomic system.


The political and economic legacies of the dictatorship have a significant effect on the PMP. The human rights abuses, together with the imposition by non-democratic means of a new constitution and a new economic system, generated a deep ideological schism in the electorate. The two party coalitions formed as a result of the 1988 plebiscite mirror the sharp ideological divide regarding the legitimacy of the political and economic systems. This translates into a bimodal ideological distribution over socioeconomic policies. In this context, Chile has a very orderly policy debate, which delves into perfecting the new economic model to increase efficiency and competition, put in place sophisticated market regulations, and increase social protection. In the context of consensus politics in a cooperative PMP with a status quo bias, this implies slow but permanent policy reform without policy reversals. The Chilean policymaking process is less transparent and with greater policy spatial fragmentation than previously thought. Despite Chile’s transparent Congress as its main policymaking arena and a state bureaucracy with low corruption levels, some institutional and cultural features increase the obscurity of political actors’ moves. Three main determinants drive these results: i) opaque internal workings of political parties, ii) inadequate conflict of interest and campaign finance legislation, iii) increasingly complex productivity-enhancing policies, and iv) strong ideological biases of the electorate and policymakers. The main negotiating arenas are Congress, the Executive branch, and political parties. Though political parties play a fundamental role in Chilean politics, their internal workings are poorly regulated. Thus, the selection of party leadership and candidates to public office occurs in backroom negotiations rather than through primaries. The low competition of the binominal electoral system increases opportunities for politicians and parties to defend private interests without paying electoral costs. Chile’s high aggregation of socioeconomic actors reduces the number of negotiating parties, thus facilitating inter-temporal policy transactions. At the same time, however, high aggregation in low-transparency arenas facilitates collusion to defend private interests, therefore facilitating capture of political parties´ policy preferences. The complexity of policy issues makes it harder for the electorate to understand the benefits and costs of alternative policy reforms, reducing the visibility of political actors’ moves. When the electorate is highly ideological, politicians can further obscure the policy debate by ideologically manipulating the public policy analysis.


The lower transparency of the policy debate partly explains the variance in policy outcomes since 1990. First-generation structural policies have high social visibility, as they are easily understandable and their effects are easily measured. In a decisive, status quo bias PMP with a new socioeconomic model, first-generation reforms were easily agreed upon, but when efficiency-enhancing reforms became necessary to maintain high levels of productivity growth, the system was unable to deliver. Education reform presents a clear example. Education policy in the 1990s was focused on increasing coverage and received broad political support. When the system achieved high coverage levels, the policy focus shifted to increasing quality, but there was no technical agreement on how to achieve this. The public could not understand the determinants of low performance of the system, and any reform implied costs to many private actors. In that context, for more than a decade it was not possible to reach an agreement on education reform. The fragmentation of the policy space is also a limitation in the Chilean PMP. In the past 30 years, Chile’s bureaucracy has undergone substantial reform, but it is still too rigid and fragmented across ministries. There are outdated and inadequate horizontal coordination mechanisms, and at the center of government there is low coordination capacity (Aninat and Rivera, 2009). This is compounded by coalition politics. Coalition governments lead to assigning different ministries to different parties. This, combined with the Weberian Chilean state, leads to fragmented policymaking in reinforcing ministries’ silo effect. Different parties have different social constituencies and are closer to different social actors. In a party system with low transparency regarding the inner workings of parties, some social actors are able to capture policy in the different ministries, fragmenting the policy space with strong negative cross effects. The negative effects of lower transparency and fragmented policy space are higher for more complex, less visible policies. As Chile develops and high rates of productivity growth depend more on policies with higher complexity, the Chilean policymaking process has become increasingly less effective. Given the structure of the economy and the historical configuration of socioeconomic actors, Chile’s main interests are big business, unionized labor, and strong ideological divisions. This leads to an equilibrium in which organized labor lobbies the Concertación parties and governments for higher social protection—higher public spending, labor legislation, and a fully public education system. On the other hand, big business, through CPC, major economic


conglomerates, and right-wing think tanks lobby the Alianza parties and the government to maintain the mostly unregulated economic system and limited government. The status quo bias of the PMP, together with the spatial fragmentation of the policy areas, has led to the maintenance of a small state, with low social protection, and a high unregulated economic system with a few areas dominated by organized labor—CUT, Trabajadores del Cobre, ANEF, and Colegio de Profesores—leading to unnecessary labor-market rigidities, such as no part-time regulation and high firing costs, disproportionately high wages for CODELCO’s workers, and gridlock in education reform.

6. Six Policy Episodes: Success, Failure, and Mixed Results in Policymaking for Productivity We have selected six policy reform episodes that have had an impact on productivity through efficiency gains or innovation. Our selection is somewhat arbitrary. However, they correspond to the kind of policies the WEF identifies as crucial to enhance productivity in a middle-income country. Two of the cases correspond to success stories in which reforms were approved and implemented: the establishment of the antitrust tribunal and a major civil service reform, both in 2003. Two selected “unsuccessful” cases are the education reform and a new corporate governance scheme, while two mixed cases are related to the royalty to mining firms and an innovation in the institutional framework. 6.1 First Successful Reform Episode: Public Administration Management System In 2003, Chile undertook a crucial move in terms of improving the criteria for selecting senior civil servants. That year, a new appointment scheme based on an open competitive and meritocratic selection process was put in place. By introducing a significant element of the modernization of the state, this reform was expected to be TFP enhancing. The old civil servant appointment system: Historically, Chile’s development was led by a natural resource-intensive strategy, with entrepreneurship characterized by rent-seeking behavior that, as such, was not skilled labor-intensive. This feature forced the State to absorb the excess of semi-skilled labor that the educational system was producing. Over time, the public sector federations gained strength and concentrated on ensuring lifelong employment for public sector workers. The National Federation of Public Sector Workers, ANEF (and similar organizations in


the health, municipal, and other branches of the public sector) would act to maximize job security and salary increases. The pressure of the ANEF focused on job stability and salaries based on seniority at the expense of performance. Also, the vacancy filling processes in the public sector were restricted to existing public sector workers. As such, there was no competitive selection process that included both internal and external candidates. Furthermore, the selection processes were only open to candidates from the directly lower level. In response to these hiring restrictions, and given the requirement for skilled professionals in the ministries (which were scarce given the selection process described above), a fixed-term hiring system with one-year contracts developed over time. This mechanism permitted ministries to hire more skilled professionals. However, ANEF opposed this hiring system, alleging that it failed to provide adequate employment conditions for workers. Filling public sector vacancies through discretional fixed-term contracts fostered a situation whereby the governing political parties put their own people into government jobs. This led to filling public sector vacancies with people that did not always have the requisite skills (actually, experts in politics not in policies). The 2002 corruption scandals: By the end of 2002, a vast corruption scandal erupted. The scandal was known as MOP-Gate, because it involved the Ministry of Public Works (MOP) and Gate, a private company that provided services to the ministry. The scandal was about cash overpayments to certain senior civil servants (including the minister) within the MOP using a complex structure of ad hoc contracts with private consultants and funds diverted through public universities. The case made evident to the public an old informal practice, namely, to finance the incumbent political parties through “budget reserved spending” and other informal mechanisms. Besides, it revealed the existence of a strong internal machine appointing a large number of senior civil servants based exclusively on political criteria rather than on merit. This high-profile corruption case resulted in a political crisis that opened a window of opportunity to discuss two major issues. The first was incumbent interventionism. Some 3,500 civil service positions were reserved for people appointed directly by the Executive (Boeninger, 2007). This discretionary power led to a large number of appointments based on partisan criteria and patronage. Moreover, the government had at its disposal a large amount of “reserved expenses” not subject to accountability. The second issue was the relationship between money and politics. The abovementioned problem underlies another issue related to political party


funding. Political parties are funded by contributions from members and by a state subsidy during election periods. However, they must also fund themselves between elections. One way of doing so is by putting their members in public sector positions that allow them to finance their political activities. Therefore, party members who have been appointed to public sector jobs are more dedicated to politics than to their jobs. Right-wing parties also face a funding problem. In their case, business groups provide funding in exchange for legislative favors. All of the above is compounded by a low level of supervision of political party spending and funding, and therefore impunity is the norm. Identifying the relevant actors: Given this situation, President Lagos rapidly brokered an agreement with the leader of the opposition, UDI president Pablo Longueira. They announced the creation of a special anti-corruption commission headed by the Chilean representative of Transparency International. For the opposition, the situation was an opportunity to force state reform and modernization. Opposition leaders were open to discussing the political financing mechanisms because the modernization of the state involves the financing of political parties (Boeninger 2007). This position was publicly supported by SOFOFA in its annual meeting. As a result, the government has the initial support for launching formal negotiations with the opposition. These negotiations were led by Minister of Interior Jos茅 Miguel Insulza and Finance Minister Nicol谩s Eyzaguirre, who established a commission with leaders of the four Concertaci贸n parties and the two Alianza parties. Right-wing think tanks were also involved. In fact, a salient feature of this reform was the role played by the most influential independent rightwing think tank, CEP. Indeed, since the end of the 1990s, CEP has been working on a new agenda for modernizing the state. It edited two large academic volumes (2000, 2002) with a detailed set of proposals put forth by a wide range of authors, including some government officials. The proposals served as the basis for the negotiations. In January 2003, a political agreement between the government and the Alianza was publicly announced as a commitment to advance a three-point agenda for state modernization, enhanced transparency and accountability, and growth promotion. According to Boeninger (2007), bundling this three-point agenda was a way to confront the natural resistance of public servants unions to the introduction of a system of meritocratic appointments. The final outcome was a law which created a Senior Public Appointments System. A technical and independent civil service appointment council (Consejo para la Alta Direcci贸n


Pública, CADP) appointed by the President with Senate approval is in charge of selecting candidates for the highest public sector jobs through a competitive and meritocratic selection process. The council presents a short list of three or five candidates to the President for final selection. The law was a significant and necessary step in the right direction for avoiding corruption and reducing inefficiency. However, it did not cover all public sector vacancies, and some public services were explicitly excluded. Some 875 vacancies should be filled in this way, while nearly 700 continue to be filled at the exclusive discretion of the president. The corruption scandal also produced a sweeping reform of campaign finance. Up to that point, the financing of political campaigns was an obscure and opaque process. Candidates would not disclose where they obtained the money for the campaigns or how much money was spent on campaigns. With the 2003 reforms, the process became more transparency. Spending caps were introduced—although there are no enforcement mechanisms and no fines for those who violate them—together with public financing for campaigns. Candidates are reimbursed money based on the number of votes they received. Thus, candidates have an incentive to campaign actively, but a high threshold exists for those who cannot borrow money for the campaign and pay later. In addition, a complicated mechanism of public disclosure of private contributions was adopted. Small contributions can be made public at the discretion of the candidates. Larger contributions must be kept secret (but are known to the contributors, candidates and, in theory, to the Electoral Service, the government agency in charge of running elections). Very large contributions must be made public. Following implementation of this reform, the system has worked much better with respect to public funding but not as well at generating transparency with respect to private contributions. At least the issue of campaign finance has been squarely put into electoral campaigns and public debates. However, the reforms did not affect the internal operations of political parties. Public financing is restricted to campaigns and does not cover the daily operations of political parties. Parties need not disclose how they raise operating funds. Thus, the system allows for opacity in the financing of political parties. The political economy of state reform: Several elements of our model are useful in explaining the final outcome of state reform. First, an exogenous shock such as the corruption scandals of 2002 made the problem highly visible to the citizens. This is why the scandals resulted in a political crisis. Since the government was forced to recognize the existence of such


bad practices, there was room to negotiate and modify the historical imbalance. In other words, the crisis opened a window of opportunity. Second, at the time of the scandals, a detailed technical proposal of state reform (including civil service reform) had already been presented by the influential CEP. More importantly, a high level of technical consensus was reached across parties, including some leading civil servants from the government who assisted in preparing the document. In a context of crisis, this technical consensus, coupled with a proposal ready to be implemented, was a precious asset to facilitate coordination among political actors. The CEP proposal was the cornerstone of the political reform implemented subsequently. Likewise, it was a useful benchmark for disciplining political parties in the negotiations. Our interviews with a broad range of first-line policymakers revealed that senior officials in the Lagos administration met regularly with CEP board members well after the corruption scandals. This was a case of bargaining among few actors in a repeated game scenario. Additionally, reaching an agreement for civil service reform was aided by the fact that the outcomes and degrees of implementation of the reform were easily measurable. Third, the negotiations between the government and the opposition had several exchange currencies to reach an agreement. On the one hand, the opposition had the right to demand a reduction in the Executive’s discretion to appoint civil servants as well as a reduction in the scope of the use of so-called “reserved expenses.” For its part, the government raised the issue of the lack of public funds for financing political parties. It was argued that the right-wing parties relied on contributions from large companies and that this financing system was very opaque. Given this, both sectors had incentives to negotiate the terms of a final project. Certainly, the opposition could have tried a one-shot gain strategy in terms of refusing to negotiate with the government and thus expanding the political crisis. This could have resulted in increasing its chances of winning in the next elections. However, this strategy is less attractive in a system with repeated games and only two actors involved (Concertación and Alianza). Moreover, from a long-term perspective, it was in the interest of the opposition to stop government electoral intervention. The final outcome was mixed. It was an important first step, but many issues remain to be resolved. Many but not all senior positions should be appointed by the CADP, and the campaign finance reform failed to consider the internal operations of political parties.


6.2 Second Successful Reform: Episode: The Antitrust and Pro-Competition Tribunal The enactment of a new antitrust framework in 2003 was an example of a productivityenhancing reform, since it improved the efficiency of the market. The new institutional framework amended substantially the existing regulatory body. Interestingly, the reform did not face major opposition when discussed in Congress. As a matter of fact, it took less than one year from its presentation until its final enactment into law. The project updated the antitrust legislation, which was out of synch with both the new economy and the country’s accelerated economic growth. First it created a truly independent framework of prosecution and resolution. Previously, both functions were somewhat mixed. Second, it endowed both the Prosecutor’s Office (Fiscalía Nacional Económica, FNE) and the Resolution Court (Tribunal de Defensa de la Libre Competencia, TDLC) with highly qualified technical members. Third, in the new framework, the power to fight against monopolistic behavior was vastly improved, for example, by defining some of the standard antitrust practices as well as by significantly increasing the fines in order to dissuade such practices. The old antitrust framework: As part of an economic program to roll back the planned economy and its state-owned companies inherited from the previous government, in 1973, the Pinochet dictatorship passed a law creating a tripartite institutional antitrust body. It was composed of a National Economic Prosecutor Office (Fiscalía Nacional Económica, FNE), a special antitrust tribunal (the Antitrust Commission), and a number of regional advisory Preventive Commissions in charge of ex-ante consultations on antitrust matters. The FNE was a lawyer appointed by the president. His office was in charge of investigating and enforcing findings on cases related to antitrust controversies. In principle, the FNE office was independent. However, its budget and its supervision were dependent on the Ministry of Economy. Until 1999, the FNE did not experience any significant change. That year, a law was passed which strengthened its powers, doubled its human resources up to 60 people, and raised their salaries. In 2001, the Prosecutor, Pedro Mattar, a well-respected lawyer who served as member of the Antitrust Commission (1994–1996), reorganized the office into three main enforcement departments, a Legal Department responsible for conducting investigations, an Economic Department acting as advisor for legal department during the investigations, and a Regulated Markets and Technical Analysis Department.


In addition to the FNE, the 1973 law defined a particular institutional body called the Preventive Commissions (Comisiones Preventivas). These were regional consultative organs in charge of answering questions (ex ante) dealing with activities that could restrict competition. There were 11 Regional Preventive Commissions and a Central Preventive Commission for Santiago and other cases involving more than one region. The Central Commission was composed of a representative from the Ministry of the Economy, a representative from the Ministry of Finance, two university professors, and a representative from the Neighborhood Associations. Commission members did not receive any pay. The third body of the antitrust framework was the Antitrust Commission (Comisión Resolutiva). The Commission was a special court whose main function was to decide cases brought by either the FNE or private complainants. It applied sanctions, including fines, penal duties, and business dissolution and restructuring. The antitrust body was under the supervision of the Supreme Court and was chaired by a judge appointed by the Court. The remaining members were two Chiefs of Service from the Economy and Treasury Ministries and two representatives from the universities (deans of the Economics and Law Departments). As in the case of the Preventive Commissions, the members of the Antitrust Commission did not receive any pay. Indeed, none of these Commissions had a budget. Moreover, while they were supposed to be independent from the Prosecutor’s Office, both of them relied upon a Prosecutor’s Office Secretariat, which provided them technical and administrative assistance. The new framework: None of the tripartite antitrust bodies were fully independent. On the one hand, the Executive appointed their members. On the other, the Preventive and Antitrust commissions relied on the Prosecutor’s Office. This was a serious shortcoming, potentially hampering the necessary impartiality required for efficient performance of the antitrust body. In addition, both the Preventive and Antitrust Commissions relied on unpaid members whose technical skills for solving antitrust controversies were far from adequate. Finally, the antitrust framework had not been updated for 30 years. In 2003, a new law (#19.911) passed to reform the framework accomplished two goals: it strengthened the antitrust policies and it replaced the Antitrust and Preventive Commissions with a new independent and technically based Antitrust Tribunal (Tribunal de la Libre Competencia, TDLC). Main actors involved: As confirmed to us by inside actors, the UDI president, Senator Jovino Novoa, was the leading voice of the Alianza during the debates. Novoa was advised by


his personal counselor, Heidy Matthei, as well as by the economic and legal staff of LyD, including Axel Buchheister, an expert in constitutional and legal issues, and María de la Luz Domper, a senior economist and an expert in antitrust matters. According to our interviews, the advising process was so meticulous that Novoa’s team “reviewed word for word every single sentence proposed by the government during the informal negotiations.” The FNE was represented by its director, Pedro Mattar, its deputy director, Enrique Vergara—who currently serves as the National Prosecutor—and its technical staff. According to interviews, a significant part of the informal negotiations between the opposition and the government were carried out at the FNE headquarters. On the government side, the negotiators were the Minister of Economy, Jorge Rodríguez, a member of the PDC, and the Vice-Minister Alvaro Díaz, a socialist, with the support of his technical and legal staff. The Senate Economy and Finance Commissions comprised senators representing all parties. Several business organizations were also active in the debate and discussions. During Senate commission debates, SOFOFA was represented by its legal counsel, Claudio Undurraga, while CPC was represented by its CEO, Carlos Urenda. SNA was represented by its General Secretary, Jorge Garcia. Finally, the Comisión Defensora Ciudadana, with its president, Francisco Fernández, a member of the Socialist Party, represented the interests of consumers and citizens. Critical issues and objections to the project: The former government project was unanimously supported by the two right-wing parties as well as by representatives of the business sector. However, several critical issues were raised. Most of the legislative debate occurring in 2002-2003 centered on these issues. From our review of the legislative debates as well as from interviews with representatives of LyD, we identified eight main controversies. First, the judicial goal to be protected (Article 1 of the Law). The former government project intended simultaneously to protect competition, consumers, and the efficiency of the market. This was opposed by actors close to right-wing parties, particularly in relation to consumer protection. It was argued that defining multiple goals could be misleading and discretionary and that consumer protection should be the subject of a different law. This opposition was supported by LyD (Informe Económico 133, October 2002), RN, and UDI, and the business organizations CPC and SNA. On the other side, the Comisión Defensora Ciudadana supported the principle of defending consumers’ interests. In the law passed in 2003, the tripartite goal was abandoned. Instead, the law favored a somewhat vague objective, namely,


“promoting and defending free competition� without clearly specifying what that meant or, for that matter, defining free competition. This ambiguity has been recognized as a problem by some of the members of the TDLC. Second, the real independence of the TDLC (Art 8). The former project stated that the PDLC would be composed of five members with three-year terms elected as follows: one Supreme Court justice elected by the Supreme Court; one economist and one lawyer with relevant experience in antitrust matters appointed by the ministers of Economy and Finance, respectively; two scholars (one economist and one lawyer) selected by the Consejo de Rectores, an organization comprising all traditional public and private universities. CPC, SOFOFA, LyD and the two right-wing parties argued that this formula did not grant the TDLC independence from the President. In the same fashion, LyD and SNA criticized the short term in office (three years) of the members of the Tribunal. The final agreement reached was as follows: the president of the Tribunal, a lawyer with at least ten years of experience in competition law, would be appointed by the President from a list of five nominees supplied by the Supreme Court after a public competition open to all interested qualified candidates. The other members (two lawyers and two economists) would be appointed as follows: one lawyer and one economist chosen by the President from a list of three nominees submitted by the Council of Governors of the Central Bank and selected through a public competition, and the other lawyer and economist would be appointed directly by the Central Bank from candidates selected by the same open and public competition process. The appointed members serve six-year terms and may be reappointed. Tribunal members receive a fixed salary plus a fee based on the amount of work done. The Tribunal must meet at least twice per week. The TDLC has his own budget and staff. The opposition won this battle, as the Executive agreed to limit its influence over the process of appointing Tribunal members. Third, the excessive amount of fines (article 17). The former project eliminated the penal sanctions for antitrust practices (a provision not invoked since 1973). This modification was broadly accepted by all social actors. However, the draft bill sent by the President had initially proposed to heavily increase fines levied by the Tribunal from a maximum of nearly $230,000 up to $15,300,000 (30,000 Unidades Tribunales Mensuales, UTM, some US$1.5 million). In addition, these fines could be issued against a company or its board of directors and management staff. This point was opposed by SOFOFA, CPC, and SNA, as well as by RN, UDI and LyD.


They considered the new fines to be excessively high and disproportionate to the damages caused. The opponents also criticized the fact that the fines would be added to the Tribunal’s budget rather than distributed among the economic actors affected. In addition, they disagreed with the criteria of extending the fines to management and the board of directors. The Lagos administration reckoned that the fines should be sufficiently high to be dissuasive. Moreover, the gravity of the crime had to be sanctioned regardless of the economic damages involved. In the law enacted in 2003, the maximum fine was set at $10,200,000 (20,000 UTM). However, the provision according to which both the management and board of directors could eventually be sanctioned was retained. All in all, the opposition obtained a point and the government kept a relevant part of its original proposition. Fourth, defined examples of antitrust practices (Art 3). The former project explicitly described three antitrust practices to be sanctioned: explicit or tacit agreements (collusion) on prices, quantities, or market shares; the abusive exploitation of a dominant market position, and predatory practices designed to increase or reach a dominant market position. The UDI, represented by Senator Novoa, was contrary to such a “vast and vague” definition 3 of the type of antitrust conduct included in the original Article 3. The argument was that these definitions could be misleading since they were not necessarily the manifestation of antitrust practices. Not surprisingly, this was the same argument made by the economist and legal experts from LyD (LyD, “Informe Económico #133”, October, 2002). The law enacted in 2003 kept these three examples of antitrust behavior and introduced some very minor changes. Thus, the Alianza opposition did not obtain the change that it was demanding. Fifth, strengthening the powers of the Prosecutor. Aside from the changes introduced in 1999, the project gave new powers to the National Economic Prosecutor, including the authority to sign agreements with domestic agencies and foreign entities, the right to take depositions from judicial and natural persons, asking other public entities “secret” information relevant to the investigation. Both the SOFOFA and the SNA opposed these changes. However, in the end, the previous clauses were retained in the final Law of 2003. Sixth, the independence of the Prosecutor. CPC, LyD and right-wing parties advocated for a designation mechanism that would grant the independence of the Prosecutor. The initial


See Jovino Novoa intervention at the Senate on September 3rd, 2002. In Historia de le Ley 19911, pp. 2592-2594.


project gave the president the power to appoint the Prosecutor with Senate confirmation. In the end, no changes occurred and the previous clauses were kept in the law enacted in 2003. Seventh, the advisory role of the TDLC. There were disagreements on granting the TDLC an advisory role. It was argued that the TDLC should solely serve as a dispute-resolution institution. That position was supported by SOFOFA and SNA. In addition, LyD was contrary to the idea of dissolving the preventive commissions that existed in the previous framework. None of these objections was finally incorporated in the law. Eight, imposing costs on the claimant in case of absolution. SOFOFA and LyD claimed that, in case of absolution, any natural or judicial person that accused a company of antitrust practices should be required to pay compensation or a fine. The argument was that this was necessary to avoid the presentation of antitrust suits that lacked merit. In the end, the law enacted in 2003 did not include this idea. Summing up, some interesting observations can be made. First, the general idea of the project proposed by the Lagos administration to create a new Antitrust Commission (TDLC) was broadly and rapidly accepted by the opposition and business associations. Yet, there were discrepancies and differences related to specific articles of the proposed bill. Second, the criticism and objections raised by Alianza parties, primarily UDI, were roughly the same as those raised by business associations. In many aspects, UDI and the business associations acted as a single block. This interesting observation was especially true in the case of the three more problematic points raised by opposition parties: the judicial good to be protected, the independence of the TDLC, and the size of the fines. Interviews with right-wing actors (both from the political parties and the think tank LyD) who directly participated in the process confirmed that these were the three most important issues for them. On these three points, opposition parties extracted important concessions from the Lagos administration in the final law enacted. However, their remaining observations (points 4 to 8 of the above list) were not truly taken into account in the final law. It is as if the opposition parties were willing to sacrifice points 4 to 8 in exchange for the government’s concessions on the first three aforementioned provisions. The political economy of competition policy: From a PMP perspective, three elements of this model are in play. First, the existence of a reduced number of actors coupled with, second,


continuous negotiations and third, technical discussion and consensus. The outcome has low visibility but high impact to some relevant vested interests. The final law was the achievement of a continuous policymaking process that started in 1999 and involved a reduced number of actors. At that time, Law #19.610 was enacted, increasing the powers and responsibilities of the Prosecutor’s Office (FNE) in antitrust matters. Since then, the government and the opposition agreed on the creation of a technical commission to reach consensus around the creation of an independent antitrust body, which finally resulted in the TDLC. From a technical perspective, the negotiations were mainly carried out by the LyD, the Ministry of Economy and its technical and legal staff, and the Prosecutor’s Office and its technical and legal staff. In the words of the Minister of Economy, Jorge Rodríguez, the project was “the natural outcome of the process started with Law 19.610.” 4 This explains why the reform did not face major opposition in Congress. It took only slightly more than one year from the government’s former proposal until its final enactment as law in 2003. Opposition parties immediately approved the first version of the draft presented in 2002 in its general aspects. The legislative discussion carried out in two commissions of the Senate (Economy and Finance Commissions) was mostly focused on the specific article points that we have reviewed. In our PMP model, visibility is a crucial condition for successful outcomes. Yet, reforming the antitrust institutions had low visibility. It was not preceded by an emblematic case or a public scandal. It involved a technical discussion far away from the public interest. The project was quite relevant for vested interests in the right-wing coalition, large economic groups. In principle, a reform strengthening the power of the antitrust body could be seen as hampering the particular interests of major economic groups. Thus, one would expect the right wing to oppose such an initiative. As seen, this was far from being the case. Not only was the former project largely accepted in general terms, but also its final approval in Congress was extremely expedient. How can we explain this success story? The initial equilibrium was bad for both the opposition parties and the business associations. Thus, there was a large win set for reaching agreements with the government, because the National Prosecutor’s powers were increased in 1999. 5 This made it necessary to counteract such a power by granting the independence of the Resolution Body (TDLC). This 4

Boletín nº 2.944-03, 08/13/2002, p. 4. Informe de la Comisión de Economía. Our interview with first hand actors from LyD reveals that this theoretical point was indeed the main reason for supporting the project.



explains why the right-wing parties and the business associations were so persistent and allied in their claims for more independence in the selection of the members of the TDLC. As previously seen, this was a battle they won. In the same fashion, although agreeing with the elimination of the penal sanctions, the opposition parties and the business sectors were against the “very high” fines established in the former project. However, in the end, they obtained a reduction in these maximum fines. They were also successful in excluding consumer protection from the first article of the project. Instead, the law defined a very broad goal. Because of this opacity, one could argue that the final outcome favored the companies. An alternative way to see this is an information approach. Since 1999, the benefits from such a reform were visible for anybody, independent of their ideology. This is probably the reason why after this event, the government, the Prosecutor and LyD began a negotiating process on reforming the TDLC. The “bad” initial equilibrium was the catalyst for ending an asymmetrical situation that had blocked the reform. In the past, although there was agreement that the project could be socially beneficial, the benefits to the right-wing parties and the companies were not sufficiently clear. The Chilean antitrust scheme relied on a framework that, despite its evident imperfections, did not truly affect the economic interest of the business class. It was in the interest of the right-wing parties “to exchange” more regulation for “less discretionary power.” This is why everybody declared their general agreement with the former government’s idea. What followed was to pursue negotiations, step by step, on the most relevant articles we have already described. 6.3 First Unsuccessful Policy Episode: Education Reform The quality of the educational system became a central issue in the policy debate at the beginning of the Lagos administration. During the 1990s, the main education policy challenge was to increase coverage for high school and tertiary education. By the end of the decade, that goal was mostly attained, refocusing national attention on the poor performance of Chilean students on international educational attainment tests (PISA and TIMSS) and on the overall quality of the education system. Indeed, Chile lags behind in terms of its scores in international standardized tests, even after controlling for its GDP per capita. In tests such as PISA or TIMMS, Chile’s scores are about one standard deviation below what is expected in countries


with similar levels of economic development. According to the WEF Global Competitiveness Report in 2008, Chile ranked 86th, 106th and 110th out of 134 countries in terms of the perceived “quality of “the educational system,” “math and science” and “quality of public schools,” respectively. This situation contrasts sharply with Chile’s good overall Global Competitiveness Position (28th). By contrast, in 2000, Chile ranked 43rd and 47th in terms of the two first aforementioned indicators. The policies needed to increase quality were highly technical, involving better regulations, supervision, and accountability, better teacher training, and higher public spending. These policies affect myriad private interests and have high ideological connotations. The educational system and the main policy reforms since 1990: The educational system is based on vouchers as a mechanism for financing both municipal and private, subsidized schools. The objective was to generate competition within the system and increase the ability of parents to match their preferences for specific values or pedagogical approaches to schooling. The system has three types of providers: municipal—public—education (45 percent of all students), a private subsidized system (48 percent) and pure private schools (7 percent). Both municipal and private subsidized schools receive a per-student subsidy of about US$70 per month. The management of municipal schools depends on each municipality. Municipal mayors do not have authority over hiring, firing, and setting teacher salaries and contracts. While public elementary schools are free by law, if the school community agrees, the municipal high schools can levy “voluntary fees,” although in practice, only about 15 percent do this. Private entities run subsidized schools. These include foundations, religious organizations, individual entrepreneurs, and companies known as “sostenedores.” Aside from the per student state subsidy (voucher), they can charge an extra fee. About 50 percent of private voucher schools remain free, and most schools charge only minimal fees, called financiamiento compartido (Cox, 2004). When democracy was restored in 1990, the government faced a number of challenges. The incoming center-left government acquiesced to deadlock constitutional provisions left in place by the outgoing regime (Boeninger, 1997; Zaldívar Larraín, 1995). In addition, three days prior to the end of the military regime, Pinochet passed laws that required supermajorities in order to pass reforms. Among those last-minute laws, the Organic Constitutional Law of Education (LOCE) constrained the new government’s ability to alter the educational model and


legal structure (Cox, 2004). Because of its origin, the LOCE became a powerful symbol of both the outgoing regime’s imposed educational model and the new government’s tied hands to tear down the “neoliberal” economic model. Among the reforms deadlocked by the LOCE was the school voucher system. More importantly, the LOCE required the state to ensure access to education but did not give it the responsibility to ensure that everyone received a high-quality education. During the Aylwin administration, the government worked with teachers’ unions to strengthen job security. The Estatuto Docente (Teachers’ Statute) defining employment immobility was hailed as a reform that gave justice to teachers, a group whose opposition to Pinochet cost them dearly in salaries and job security. Two problems—voucher schools oversight and teacher accountability—became the issues over which all educational debates were fought in subsequent years. On the one hand, the Concertación regularly attributed the deficient educational results to the lack of government oversight power over privately run schools. On the other, the opposition blamed the Concertación for not adopting mechanisms to increase teacher accountability. The voucher system was partially modified under Aylwin. The Ley de Financiamiento Compartido (Shared Financing, in 1993) allowed voucher schools to charge fees to complement the government subsidy. That reform, heralded as an opportunity to improve the quality of the education received by children, helped worsen the already unequal access to education. The second Concertación administration (Eduardo Frei, 1994-2000) focused on reforms designed to expand coverage, improve infrastructure, and provide parents with school performance information. Although tests had been administered since the mid 1980s, the results were only made public after 1990. At first, results were tabulated by cities rather than by schools. Thus, parents had limited information on the quality of the education their children received. The full-day school reform (Jornada Escolar Completa) became a symbol of the Frei’s approach to educational reform. Rather than take on the vested interests of teachers’ unions or voucher school owners (“sostenedores”), Frei implemented reforms that increased educational spending but failed to improve accountability. National test scores began to show that increased spending had failed to improve performance. The third Concertación administration (Lagos, 2000-2006) introduced reforms to improve teacher accountability through a voluntary teacher evaluation program tied to salary


incentives. Yet, there was no mandatory evaluation given the opposition of teachers’ unions and the reluctant support such initiatives found among Concertación legislators. The government failed to push legislation to strengthen its oversight power over private voucher schools. Reform initiatives were blocked by Concertación legislators who opposed introducing more accountability tools to improve teacher quality and by Alianza legislators who opposed a stronger regulatory framework. Because reforms required supermajorities, the government opted to improve access to information, introduce better quality measurements, and increase funding for infrastructure, salaries, and subsidies to needy communities. Shortly after taking office, the fourth Concertación administration (Michelle Bachelet, 2006-2010) was hit by a surprise national secondary student protest. The students opposed unequal access to education. Since the restoration of democracy in 1990, voucher schools had grown from 32 percent of all students in 1990 to more than 40 percent in 2005 (Brunner et al., 2006). As a response to the protests, in June of 2006, Bachelet appointed an 83-member Educational Quality Advisory Council and charged it with proposing a comprehensive education reform to improve quality and reduce unequal access. The advisory council issued a nonunanimous report in December 2006. Contentious issues divided council members along ideological lines. While the majority of council members—sympathetic to student demands— favored a requirement that would prohibit voucher schools from discretional selection of students (a practice that allowed rejection of high-risk students), a minority of members sought to promote competition among educational providers. The workings of the council are invaluable as a laboratory for understanding the main issues and ideological positions of the various political and social actors involved in the reform. Relevant Actors and Issues of the Chilean Educational PMP: In December 2006, the Council produced a large policy document with an assessment and recommendations. 6 Because the Commission was composed of 83 individuals from across the political, institutional, and technical spectrum, it represents a good sample of the main social actors involved in the educational debate, their ideological and technical positions, and their points of agreement and divergence as well as their main arguments.


The 252-page document is available at :


Table 3. Presidential Education Advisory Council (2006) Constituencies represented Congress and government officials University representatives Private school representatives Teachers representatives Student representatives Think tanks Municipalities High school students Church Other TOTAL Source: Authors’ compilation.

Number of members 4 16 4 6 16 7 7 13 3 7 83

percent Total membership 4.8 19.3 4.8 7.2 19.3 8.4 8.4 15.7 3.6 8.4 100

Think tanks: There were four think tanks represented on the Council. LyD exerted a strong influence on right-wing legislators. CEP, the influential liberal technical right-wing think tank, has formal independence from the two conservative political parties. Partially funded by the Matte family, one of the richest in the country, CEP is generally perceived as advocating for the interest of the market-friendly entrepreneurial class. Expansiva is a liberal and highly technocratic left-of-center think tank founded by Andrés Velasco, the current Finance Minister. Its views are mostly influential within the Concertación, but somewhat unpopular among leftwing legislators. The Centro de Estudios Nacionales de Desarrollo Alternativo (CENDA) is a less influential leftist ideological think tank, but its views are popular among student representatives and also among some left-wing Concertación legislators. Teachers Unions: The Colegio de Profesores represented the interests of teachers. Composed of teachers from the (public) municipal school system and politically close to the Socialist Party in the Concertación and the PC, The Colegio de Profesores favors a fully public schools approach and opposes voucher schools. It also defends the Estatuto Docente, which grants tenure to public school teachers and determines salaries based on seniority rather than performance. Teachers were also represented by the Central Unitaria de Trabajadores (CUT). Though not directly concerned with education, CUT espoused views sympathetic to those of the Colegio de Profesores. CUT is politically close to the PS and PC. The Asociación Nacional de Empleados Fiscales (ANEF) is a strong union of public sector employees. Not directly concerned with educational topics, ANEF supports the views of the Colegio de Profesores and CUT and has a representative on the Council as well. 46

Private school representatives: The Council also had representatives from private schools. The Matte Schools were represented by Patricia Matte, a member of the wealthy family that runs the schools (and funds CEP). Matte is also a leading educational advisor for LyD. Matte schools are a successful nonprofit voucher school group. In a sense, Matte Schools represented a large number of private voucher (subsidized) schools. They advocate for school freedom to define their educational programs and to select their students. Because of their good educational results, Matte Schools are a standard bearer for those who argue that bad educational performance of schools has more to do with poor management than with a lack of financial resources. Critics maintain that Matte Schools and other quality voucher school can show better results precisely because they select the best students from among the group of low- income family students that have no choice but to attend public schools or voucher schools that do not charge an additional fee.

In fact, critics contend that the presence of Matte Schools and other

well-ran private schools worsens the already difficult situation of municipal schools in poor neighborhoods, since most good students are recruited by these private nonprofit schools, leaving the poorer-performing students behind in municipal schools. Student federations: Students were also represented on the Council. Divided into secondary and higher education students, the unions gained recognition after the “Penguin Revolution” of 2006. Among these unions are the Confederación de Estudiantes de Chile (CONFECH), the Confederación de Estudiantes de la Educación Superior Privada (CONFESUP) and the Asamblea de Estudiantes Secundarios (ACES). In general, student unions claimed the political credit for the initiative to reform the educational system. They advocated in favor of more equal access to education and equated their demands with the abolition of the LOCE law. Scholars: A number of scholars also sat on the Council. We do not aggregate them in a single body, since they are generally associated with think tanks or universities. They were invited for their technical expertise. Interestingly, most of the influential Chilean scholars on educational policy are economists. Universities: Representatives from universities were also convened on the Council. Universities can be classified as belonging to the Consejo de Rectores, and thus recipients of direct public financing (Aporte Fiscal Directo, AFI) and private universities created after 1981


that do not receive AFI. The Consejo de Rectores universities tend to share views more sympathetic with state intervention in the educational system while the latter endorse competition in the educational system. Political parties: Because of the binominal electoral system, most parties are grouped into two coalitions with parliamentary representation, the center-right Alianza and the center-left Concertaci贸n. The two other left-wing parties with no parliamentary representation are the PC and PH. Within the Alianza, the largest is UDI, while RN is closer to the center. In the Concertaci贸n, aligned in a center-left continuum, are PDC, the largest, PRSD, PPD and PS. Ideological positions on key issues: According to the Report of the Education Council, there was a relatively shared assessment on the bad quality of the educational system and on unequal access. There was also a common view on the need to increase public funds, but not on how such funds should be allocated. Positions diverged on the role of the private sector in the educational system and competition as quality enhancing. Consequently, positions were in conflict on some key ideological issues. State provision of public education: Some sectors claimed that public education means that the state provides it. Moreover, they argued that public provision is better than private provision. Teachers unions, student unions, public university representatives and left-wing parties overwhelmingly favored these views. We shall call them pro public-run education. Others argued that the system should be mixed, that public (municipal) schools should co-exist and compete with private subsidized schools. The role of the state should be in financing, but not necessarily in providing education. Think tanks, right-wing parties, private schools, and private university representatives favored these views. We will refer to them as the pro public-financed group. Scholars were divided between both groups. For-profit schools: The first group claimed that private schools could only exist if they were nonprofit. The second group, the pro public-financed group, argued that for-profit schools are inherent in a competitive efficiency-enhancing market. In between, there emerged the view that schools could be for profit as long as they did not receive public funding. Freedom to choose programs: The first group argued in favor of a common curriculum for private (subsidized) and public schools. The second group argued against on the grounds that it would violate the right of parents, and schools, to select educational priorities.


Quality agency: In line with ensuring minimum quality, the first group proposed a public quality agency. Vouchers should be conditional on the fulfillment of minimum quality standards. Opponents of this view claimed that creating such a public agency would run the risk of ruling against the freedom to select educational curricula. They opposed the standardization and centralization of the curriculum. Student selection: The first group argued in favor of prohibiting public and private voucher schools from selecting students. Schools that received government funds would be banned from selecting students. They would be required to implement blind admission policies. Opponents claimed that blind selection would limit freedom for religious schools that would normally prefer to admit students whose parents shared the school’s religious affiliation. Table 4 presents a simplified description of the positions of some of the relevant actors previously identified in terms of the aforementioned ideological issues.

Table 4. Ideological Positions on Educational Reform in Bachelet’s Advisory Council, 2006 Issue Mixed Provision of the public education

Against (-) Colegio de Profesores, UT PS, Student Unions Public universities

Undecided (+-) Part of the Concertación

For profit schools

Colegio de Profesores CUT, Student Unions Part of the Concertación

Part of the Concertación, some scholars, some universities Some scholars,

Colegio de Profesores, CUT, Students Unions, part of the Concertación Rightwing parties (UDI) Private schools (Matte schools), Libertad y Desarrollo Colegio de Profesores, CUT, Freedom to select students Students Unions, most of the Concertación, some scholars Source: Authors’ compilation. Freedom to choose programs Quality agency

Renovación Nacional Centro de Estudios Públicos Some scholars


In favor (+) Right wing (UDI, RN), Part of Concertación, Private schools (Matte schools) Most scholars, LyD, CEP, Expansiva, Private universities Right wing (UDI, RN), Private schools (Matte schools), some scholars, LyD, CEP, Expansiva Right wing parties (UDI, RN), Private schools representatives (Matte schools), some scholars, LyD, CEP,Expansiva Colegio de Profesores, CUT Students Unions, most of the Concertación, most scholars Right-wing parties (UDI, RN) Private schools (Matte schools) Some scholars, LyD, CEP

The political economy of education reform: Because the Council had more members inclined to support publicly administered—rather than publicly financed—education, the final report reflected those views. In addition, because the student unions that had triggered the protests that eventually forced the government to form the Council were vocally in favor of the pro-publicly administered view, the final report reflected that vision. President Bachelet sent legislation to Congress in early 2007 to replace the LOCE. The proposed legislation included eliminating for-profit voucher schools (voucher schools run by religious organizations and non-profits would still be allowed), the strengthening of public schools and the banning of discretionary selection. Although the proposal was rejected by the Alianza—because it went after sostenedores but not teachers—a compromise was brokered between the government and the opposition. Yet, the legislation was far from a done deal. Concertación legislators rebelled against the agreement. When Bachelet made concessions to them, the Alianza disavowed the agreement. The legislation has been stalled in Congress since. An unrelated corruption scandal in early 2008—associated with lack of oversight of voucher payments to privately run schools in Santiago—diverted attention from the debate over the new legislation. The Senate impeached the Minister of Education—appointed after the previous minister was sacked amid the student protests in 2006. Although a new Minister was appointed—the third under Bachelet—there was little room for negotiation, and the comprehensive reform was frozen in Congress using procedural tactics. The tension between a government coalition that sought a reform to undermine voucher schools while protecting teachers’ unions, and the opposition with the exact opposite interests, has effectively blocked reform. The progress made since 1990 has been achieved without undermining the power of teachers and sostenedores. Although some progress was made in educational coverage, infrastructure and teachers’ salaries in the first three Concertación administrations, there seems to be no more room for improvement unless the reforms affect the vested interests of the two most powerful groups involved in the education sector—private school owners and teachers. Since democracy was restored, the average tenure of a Minister of Education has been one and a half years, the shortest of any ministry. The constant change in leadership has undermined the government’s ability to push through its agenda and has strengthened the influence of vested interests and their lobbying capacity with political parties represented in Congress.


Despite their diverse nature, school owners have found common ground to oppose government-sponsored reforms. Private for-profit owners interested in limiting government oversight have teamed up with not-for-profit religious schools interested in maintaining discretionary selective mechanisms in favor of families that share their religious values. Nonprofit educational institutions also favor selection mechanisms, for they can promote their educational vision and discriminate in favor of students and parents who share their educational objectives. As a result, private voucher schools have formed a cohesive group that opposes Concertación governments’ efforts to strengthen the regulatory framework, increase government oversight, and limit discretionary selection. The Colegio de Profesores has blocked the government’s efforts to introduce stronger quality control mechanisms. After years of negotiation, the government introduced a voluntary self-assessment evaluation for teachers. Although there were salary incentives associated with participating in the self-assessment, its implementation failed since there were practically no costs for those teachers who failed the self-assessment evaluation. Because the components of the educational reform debate are opaque and confusing to public opinion, vested interest can more effectively block the government’s effort to bring about reform. Parents are interested in quality education, but they are not well informed as to what constitutes quality education and how it can be achieved. Public opinion polls regularly show that parents are aware that the quality of education received in average by Chilean children is not good. But they also think their children are getting good education at their schools. If parents cannot easily assess the quality of the education they receive, the opaqueness of the educational debate makes it difficult for the government to make its case in favor of a sensible reform. Conversely, when the problem is more transparent—that is, when the cause of the problem is more easily identified—the government can move a reform forward more easily. The students’ protests of 2006 showed that Chileans backed a comprehensive reform. When students protested, public opinion strongly supported the notion that something had to be done to improve the quality of education. Yet, when details were discussed, there was no agreement as to what was best. To be sure, some minor yet significant reforms have been implemented since 2006. The government was able to secure congressional support for a legislation that gave a bigger voucher to schools serving students from the poorest quintile of the population. Thus, a school located in a poor neighborhood would get more money per student than a school in a middle or


upper class locality. That reform, passed in 2007, was intended to partially correct existing unequal access to education and to provide incentives for private voucher schools to seek out students from low-income families. That reform also indirectly increased funding for municipal public schools that primarily serve low-income students. That reform was passed because it benefited different constituencies and did not directly hinder any powerful constituency. Yet, other reforms proposed have been blocked by powerful constituencies. The end of selection in voucher schools has been blocked by a coalition led by for-profit voucher schools. The end of selection would reduce discrimination against at-risk children (who are more challenging to educate as they come from dysfunctional or poor families, crime-ridden neighborhoods, etc). But it will not induce competition among schools nor will it create incentives for schools to provide better education. Yet, the end of selection is championed by left-wing Concertación parties who are ideologically opposed to privately run schools and would much rather prefer an all-public school system. Similarly, the government’s proposal to end for-profit schools responds to the growing influence of left-wing parties within the Concertación—especially during the Bachelet administration. Yet, that reform has been blocked by the Alianza, whose ideological position makes it favor private enterprise and oppose state-run initiatives. Teachers’ unions favor the elimination of for-profit schools because for-profit school owners optimize their earning by reducing costs (teachers’ salaries). Yet, for-profit school owners reckon that teachers are only concerned with their salaries but oppose any form of control over the quality of their teaching. Naturally, quality control is more efficient when conducted by for-profit voucher school owners. As a result, the Bachelet administration has found it impossible to deliver on its promise of introducing a comprehensive educational reform. After previous governments reached the limit in adopting reforms that did not challenge the vested interests of teachers and voucher school owners, the Bachelet administration found it impossible to advance a reform that could undermine the power and influence of either the teachers’ union or the voucher schools owners. 6.4 The Second Unsuccessful Policy Reform: Corporate Governance Reform Protection of minority shareholders’ rights became a concern for policymakers by the early 2000s as a potential restriction for capital market development, therefore limiting future productivity gains. A survey by Universidad Adolfo Ibáñez (June 2008) revealed that three out


of four executives believe that using inside information is a common practice in the financial market. We discuss below the failed effort championed by the Lagos administration in 2002 to bring about corporate governance reform. The reform submitted to Congress by the Lagos administration did not capture public and media attention, but it did negatively affect powerful private interests that vigorously opposed it. Left-wing political parties did not feel the need to defend the proposed reform (and use their political capital) as their electoral base did not perceive it as affecting their interests. Despite the significant positive effect of the bill over the large majority of shareholders, the right-wing opposition coalition, strongly lobbied by powerful economic interests, was able to publicly portray the reform as unnecessary red tape, killing the reform without paying any political price. The development of deep capital markets is a key element for enhancing productivity in an economy. Deep capital markets allow for the savings of risk-averse households to flow into risky, innovative projects with large potential upsides in terms of productivity gains. By contrast, when capital markets are thin, investors are left with limited options to finance their projects. In less developed capital markets, those options are usually restricted to the firms’ own funds or traditional bank lending. Firms are generally reluctant to finance risky projects with their own funds because they are not able to hedge the associated risks. Similarly, banks are reluctant to take on too much risk in their portfolios given their need to safeguard the value of their assets to honor their liabilities with depositors. However, as capital markets get deeper, a number of other financial intermediaries and products develop, allowing for the risk to be diversified across a larger number of agents and markets. In turn, this deepening widens the financing options available to the more innovative projects. A number of institutional features lie behind the depth of capital markets. Perhaps the more salient are the rate of economic growth, its stability, and the protection of property rights. In the latter domain, one aspect of particular importance is the regulatory framework for corporate governance and its enforcement. Corporations that seek financing through public offerings present a complex property structure that makes it possible, in the absence of an appropriate regulatory and supervisory framework, for some shareholders and/or managers to benefit at the expense of other shareholders without their consent or knowledge. When such events occur, the development of capital markets becomes seriously impaired or even suffers severe setbacks.


Corporate governance was not an important priority of public policy in Chile after 1990. Initially, the debate was dominated by the controversy over the privatization of public enterprises during the authoritarian regime. However, since a legal agreement was reached during the transition between the democratic leaders and the representatives of the old regime, an agreement that forbid revisiting those privatizations, the debate finally abated. But in the second half of the 1990s, a number of takeovers took place in the corporate sector, including in a few former public entities. Some of those takeovers caused a public scandal because minority shareholders denounced rent extraction. As the problems created by an inadequate regulatory framework for corporate governance have become apparent to the general public, the issue has been consistently in the public debate, and a number of legal initiatives have been discussed. Chile has a quite deep and developed capital market. In relation to its GDP, market capitalization is far above the ratios exhibited by other Latin American countries. But given the relative small size of the Chilean economy, the absolute level of market capitalization is still limited and can potentially grow substantially more, as can be inferred from the ratios achieved by other small to medium-sized developed economies. That achievement would be highly desirable to foster productivity. But to understand why such a sophisticated and technical topic, even considering its importance for economic growth, has sometimes acquired a high level of visibility, it is important to bear in mind two features of the Chilean economy. The first is its pension system. As a compulsory capitalization scheme for employees, the pension system makes the vast majority of workers shareholders of corporations. This is because the Administradoras de Fondos de Pensiones (AFP), the entities that collect and invest mandatory pension contributions, allocate part of their portfolios to the domestic stock market. The second is the manner and extent of the rent extraction that took place in some of the abovementioned takeovers. The most notorious one was known as the Chispas case, a takeover by the Spanish corporation Endesa of the former Chilean corporation Enersis. The latter was an electricity holding company that had been privately developed on the basis of Chilectra, a former state electricity company. When Chilectra was privatized in 1987, a fairly large group of former managers and employees of the company received the full amount of their severance payments, among other benefits, in the form of company stock. This group, in turn, created a company called Chispas, which had enough stock to become the controlling shareholder of Chilectra. Led by a small team that owned just a tiny fraction of Chispas shareholdings, Chilectra initiated an


ambitious and profitable process of expansion and transformed itself into the Enersis holding company, with assets that included electric companies in other countries in Latin America. This small team within Chispas, in turn formed by the former top managers of the former state company and closely linked to the authoritarian regime, positioned itself in the key decisionmaking positions of Enersis, gaining de facto control of the conglomerate. During the 1990s, the expansion of the conglomerate was so fast that eventually it became short of capital to continue growing. The small team decided then, in October 1997, without any notification to the boards and shareholders of the various companies of the holding, to sell it to Endesa España. The negotiations of the takeover were undertaken by the small team—on behalf of all the shareholders of Enersis—on extremely beneficial terms for themselves at the expense of the rest of the shareholders, using the “de facto control” they had acquired. As could be expected, the rest of the shareholders tried to reverse the operation both within the holding company and in the courts. However the legal framework for corporate governance that existed at the time was so weak that the shareholders failed in their attempts to obtain any compensation (although 10 years later the small team was fined with severe pecuniary sanctions by the courts). In 1998, the government submitted to Congress a draft law designed to regulate the takeovers, attempting to ensure that if the controlling shareholders would sell their shares to another investor, all other shareholders could benefit if they wished to sell their shares at the same price. The law was called the law of Initial Public Offerings (IPO law). The draft law prompted fierce resistance from the incumbents. The presidents of the Chilean Stock Exchange and the CPC, the strong organization that congregates Chilean businesses, expressed their opposition before Congress. They argued that when takeovers take place, controlling shareholders have the right to be compensated for the so called “control premium,” and the attempt to force that party to share the premium with the rest of the shareholders was inappropriate and would lead to more concentration of the corporate sector and, consequently, it would reduce the “float” (shares commonly traded) in the stock market, impairing its development. Not surprisingly, representatives of LyD presented exactly the same views. As those opposed to the law argued, the IPO law would not favor the interests of the incumbents and the so-called “control premium” would eventually be diluted. That may well


lead controlling shareholders to close the companies or to increase their shareholdings. As a matter of fact, that has been the trend in the last ten years. But it is also true that if minority shareholders feel they are exposed to value expropriation, their appetite for buying/holding stock will decrease. That will produce a reduction in the stock’s value, which would increase the cost of raising capital in the market, which would raise the entry barriers in the corporate sector and reduce competition and the appetite for risk and productivity growth. This is a classic example of the dichotomy between pro-market and pro-business policies. The draft law did not prosper for the last two years of the Frei administration (1998-99), since the government and the opposition could not reach an agreement. With the arrival of President Lagos, a compromise was reached after intense negotiations. The controlling shareholders were given three years to transfer their control to an interested party without the obligation to do so through an IPO. Even after that period, bilateral agreements for transferring control were allowed, provided that the share price implicit in the agreement would not exceed the market price by more than 15 percent. One might wonder why such an agreement was possible, why the opposition acquiesced to the agreement, given the existing political economy. Two possible explanations are evident. First, the visibility of the potential value extraction did nothing but increase with time, given that a number of such operations continued to cause public disgust. But although the visibility of the problem was at times high, the complexity of the solution made it possible for the powerful vested interests and their political representatives to at least mitigate the price of blocking the solutions with little political cost. Secondly, the agreement softened the eventual losses that controlling shareholders would have suffered, giving them time either to close their companies or to sell them, collecting the “control premium” at a discount. Given the political economy of these largely opaque although important economic issues, as discussed throughout this paper, only a “second-best” solution was politically available. The IPO law made limited progress in other areas where potential value extraction was possible, such as transactions with related parties or disposal of what could be considered “essential assets.” In principle, the controlling shareholder—or more generally, managers who have decision-making powers within a corporation—could extract value from the rest of the shareholders by buying or selling goods or services from outside suppliers at above-market rates or prices. Of course, these suppliers share common interests with the controlling shareholder.


Similar results could arise if the controlling shareholder (or key managers) sells “essential assets” of the corporation to related parties. In mature capital markets, this problem has been confronted through a number of board committees that carefully review the company’s operations and try to identify possible transactions with related parties at prices that differ from market values. But in those mature markets, the incentives for board members are aligned with the interests of all shareholders, since there may not be a controlling shareholder because the property is generally diluted. Thus, the committees are meant to control managers. In Chile, as in other emerging economies with income concentration, it is common to find extremely wealthy individuals who can, by themselves or through a small group of peers, control enough stock of a corporation to be able to appoint the majority of its board, and hence become the controlling shareholder. The power of this controlling shareholder is so strong that minority shareholders do not get to elect a board member or a sufficient number of members to guarantee the committees’ independence from the controlling shareholder. Hence the board and the committees are there to protect the interests of the controlling shareholder, not of the minority ones. The problems associated with transactions with related parties acquired public visibility in the late 1990s, when an international telecom holding company that was the controlling shareholder of a Chilean telecom corporation acquired an “essential asset”—the Internet service provider Terra—from the Chilean corporation at a price that was considered by many to be significantly below market value. The problem associated with transactions with related parties received further public attention during the early 2000s from a very different prospective. This time, members of the opposition parties denounced the practice of transactions with related parties in public enterprises, practices that were reducing their profitability and, therefore, were affecting the interests of taxpayers and/or of those that would benefit from public expenditure. After months of intense work between the government and the skilled Finance Commission of the Senate, another capital market draft law designed to improve corporate governance was presented to Congress. As could be expected, a compromise with the Senate Finance Commission was very difficult because opposition Senators tried to water down the regulations relative to the private sector, while members of the Concertación were cognizant that stringent requirements on the governance of public enterprises could become an issue for a number of Senators in their camp (given the power of public enterprises unions in the Concertación). However, the expertise of the


members of the Commission and their deep understanding of the damaging collateral effects of a weak corporate governance legal framework on economic growth moved them to embrace and actively improve the draft law. After these improvements, the main elements of the draft law can be summarized as follows. To ensure the appropriate control of potential value extraction through transactions with related parties, the draft law proposed the appointment of at least two independent board members, who would hold the majority in the now mandatory audit committee. This committee would scrutinize the transactions of the corporation and try to assess whether those transactions with related parties were made at market prices. For private corporations, the draft law proposed increasing the number of board members, if needed, to ensure that two independent members are elected, unless the controlling shareholder owns an overriding majority of the company stock. For public entities that did public offerings, two independent (i.e., independent from government) board members would be appointed. (It should be noted that these public companies do not issue stock; their public offerings take the form of bond issues, but bondholders do not have voting rights.) To ensure independence, various formulae were discussed, with the common feature of involving directly or indirectly the Senate, where, as discussed in the early sections of this paper, the opposition normally controls one half of the voting power, due to the binominal electoral system. As we discuss below, the draft law also controlled another form of potential value extraction known as insider trading. Once again, however, organized groups with vested interests attended the Finance Commission hearings to express their opposition to the draft law. It is remarkable to see their arguments, given the fact that they supposedly represent views from opposite ends of the political spectrum. During the Senate Finance Commission hearings, the most relevant unions of the public entities (which had placed bonds in the markets) expressed their disagreement with the draft law claiming that it would impose excessive control on these entities. This is because in a public holding company, board members would oversee each entity of the holding and, thus, independent-minded members would have a crucial role in appointing independent board members of those entities. In the unions’ view, this dual role would be detrimental to the efficiency of the entities. Some expressed fear that this would be a first step towards full privatization. In the case of the private companies, representatives of business organizations


argued that increasing the number of board members would be costly and would clog the fluidity of the boards’ operations. Although the Senate Finance Commission unanimously supported the draft law, the full Senate rejected it in the last working day of parliament of the Lagos Administration. The draft law needed a high quorum to be approved for discussion; thus, despite receiving the support of the majority, as it did, the draft law failed to clear the first hurdle needed for Senate consideration. The Senators who voted against the draft law represented regions where the more powerful public entities were more prominent. Non-elected Senators from the Armed Forces also voted against the law (those non-elected senators no longer exist). By law, the Armed Forces receive 10 percent of the total sales—not profits—of CODELCO. Therefore, like the unions, the Armed Forces are interested in maximizing production and not profits (the latter being what affects the taxpayer). Since then, there have been some unsuccessful attempts to improve corporate governance. Regrettably, the public’s interest in this pending issue has faded. As can be easily seen, all of the elements that have been highlighted as severe shortcomings of the current state of the PMP were at work: some visibility of the problem but a cumbersome and opaque solution for the median voter, the policy solution affects directly the interests of the more powerful and ideological groups, and these vested interests groups can work their way out of trouble through their influence in the political parties, since the opacity of the solution renders the political costs negligible to the ones that oppose the solution. Last but not least, another issue that surfaces periodically as a concern for the adequate protection of minority shareholder rights is the use of insider trading. A number of substantial stock trading by insiders, before essential information of a company is released to the market, has caused public debate and prompted investigations and legal actions by regulators. To be sure, the law does specify, to a great extent, the agents within a corporation that do have access to inside information. The list of such agents has been enlarged recently by subsequent legislation. However, a quite peculiar situation remains. What is prohibited is to use that privileged information when trading, but the law does not assume that that information is used even when trading is done during the window in which some relevant information about a company is known by the insiders and the time it is released to the market. To sanction those practices, the regulator needs to prove that the insider agents actually did use that privileged


information when deciding to buy or sell company stock. What makes that proof difficult is that generally the agents with inside information are active in the market, buying and selling paper. The regulator needs to prove that transactions with inside information were different from the agent’s usual pattern of trading. At the same time, these agents are generally wealthy and powerful individuals that can hire expensive and astute lawyers who could develop a public opinion campaign where they portray themselves as victims of political persecution. Following practices in mature economies, the draft law sought to end this odd situation. A rule in force in the United States, known as the swing-out profit rule, states that, if during the abovementioned time window agents conduct transactions with inside information, and the transactions become profitable or avoid losses to these agents, the sums involved must be given back by such agents. This rule was included in the draft legislation we have been referring to, but opposition think tanks, and hence the legislators in that camp, rejected the idea arguing that these agents were “market makers” and, therefore that the rule would damage the fluency of the market. Confirming the model we have developed in this paper, the complexity of the solution meant, once again, that the powerful interests negatively affected could exert a sort of veto power without incurring political discredit. In sum, the corporate governance regulatory framework in Chile is still weak. Some improvements have been made, but others are still pending. Improvements have been possible when the issues have acquired high public visibility and an expeditious solution “easy to understand by the general public” has been proposed. In contrast, when the issues, even with some public awareness, are too complex to mobilize public opinion in pursuit of a solution, vested interests have prevented change. 6.5 Mixed Success and Interdependence of Reforms: Specific Mining Tax and Institutional Arrangements for innovation Chile clearly lags behind on the “inspiration” side of the economic growth equation. Innovation and related activities appear to be among the most neglected areas. By mid-2004, a legislative initiative was sent to Congress with two objectives, to tax the mining sector and to create an Innovation Fund. The main argument behind the project was that a non-renewable resource was not generating enough assets to compensate for its exploitation and concomitant depletion. At the same time, a new institution in charge of securing a proper use for new resources was also


needed. After a year of discussion, only the tax portion of the project passed, while the institutional reform associated with the FIC is still under discussion in Congress. This case can thus be broken into two separate but related processes. First, a miningspecific tax aimed at financing innovation. Second, a major institutional reform, tied to regional preferences in the allocation system behind the tax collection, which sought to establish a state— rather than just a government—policy stressing the importance of innovation. Here, we analyze the innovation case in two separate blocks. In what follows, we first provide a historical overview. We then discuss what triggered the reform and the main political actors involved. Then we discuss the final outcome and describe the equilibrium reached. 6.5.1 First Mixed Success Policy Reform: Specific Mining Tax Between 2000 and 2004, several events triggered a debate on excising a royalty on mining extraction. On the one hand, there was a widely shared view that several large MNCs mining firms were not paying income taxes. The companies avoided paying taxes by using a simultaneous application of an accelerated depreciation mechanism 7 and the interest discount that can be paid to their headquarters for debts incurred in the initial financing of their investment (Boeninger 2007: 185). On the other hand, several political actors argued that firms were not paying for rents that this scarce and non-renewable resource might generate (Ricardian rents). For historical and political reasons, this second argument was very popular with the public. Although the reform was sent in 2004, there were several previous initiatives aiming to tackle this extremely beneficial tax situation for mining firms. One could even claim that those initiatives were simply corollary efforts of the widely popular nationalization of the copper industry championed by the Frei (1964-1970) and Allende (1970-73) Administrations before the Pinochet dictatorship. Copper was popularly referred to as “the salary of Chile,” and its nationalization was unanimously approved in the 1960s. Thus, even though private mining firms entered Chile during the Pinochet dictatorship, the public generally perceived that copper—if not


Most foreign investors use the Estatuto de Inversión Extranjera (FDI law, Decreto Ley 600 in place since 1974), an instrument that guarantees non-discriminatory treatment under a legal agreement between foreign firms and the state. Foreign investors have access to a uniform exchange rate, 100 percent of the firm’s property; they may instantly take their profits out of the country, and they may also choose among different tax regimes available. Any modification to the investment contract should have the agreements of all involved parties in order to protect investor interests.


all mining activities—should be state-owned or that, at least, the benefits should be captured by the state. Some of the most publicly known initiatives to tax private mining activities during the 1990s were championed by PDC Senator Jorge Lavandero, who publicly campaigned in favor of pressing large MNCs firms to pay their fair share. However Lavandero’s almost maverick cause never had the support of the Concertación governments until the issues related to tax loopholes used by some mining firms provoked public outcry. Probably, the best known case was the Disputada de las Condes mine transaction process. Its owner, Exxon, sold it to Anglo American at a high price that failed to reflect the losses claimed in their tax declarations. Moreover the transaction itself was done overseas without being subject to Chilean taxes. At the same time, since 2003 there were negotiations between the Ministry of Finance and the private association of miners (Consejo Minero) to increase taxes in this particular sector. The idea was to force firms to voluntary decline the tax benefits included in the DL600 which rules foreign direct investment. Since the Council believed that a tax increase would never pass Congress, negotiations did not advance substantially. The idea to levy a mining royalty started to gain popularity in 2003 as it became clear that the government would need to raise additional taxes to finance a major health program named Plan Auge as well to finance an ambitious social program, known as Chile Solidario, focused on the lowest quintile of the population. As Boeninger (2007) suggests, the fiscal situation was a major issue that helped build support for a mining royalty. Boeninger mentions two additional tax issues. First, the adoption of free trade agreements with the United States and the European Union would result in less tax collection. Second, Plan Auge required additional funding that was originally intended to be raised from tax increases on people with private health insurance. As free trade agreements reduced revenues and Plan Auge increased spending—without providing additional revenues—a mining tax or royalty became increasingly popular. The Lagos administration first proposed raising the value added tax (IVA) from 18 percent to 19 percent and to levy additional taxes on tobacco and diesel. Although those proposals were very unpopular, Concertación legislators agreed to support them only if the government agreed to establish royalty on the mining sector. Boeninger (2008) explains that, pressed by the entire Concertación, the Ministry of Finance sent an ad valorem tax proposal together with the creation of an Innovation Fund (FIC) in mid-2004.


The royalty-related articles were not approved in Congress as the Concertación failed to muster the needed 4/7 majority. A new version of the legislation was sent to Congress in 2005 under the figure of a specific tax which did not affect income tax laws and for which only a simple majority was needed to pass it. Given public concern about the issue just before elections and also that government credibility could be in jeopardy in foreign agencies, a rapid agreement emerged. Main Players and Arguments: Both the Consejo Minero and the SONAMI (National Mining Society) rejected the idea of royalty on mining production. Together with Sofofa and CPC, Consejo Minero and SONAMI pressed opposition legislators to vote against the initiative. The main argument to attract support from other sectors was that mining would be the first sector to be taxed. In fact, a precedent would be set for other royalties or the like to be proposed for other sectors in the future. Since modifications to the FDI Law were needed as well, it was also argued that a royalty would send a bad signal to foreign investors. However, opposition parties did not agree on the non-renewable resource argument. Some opposition leaders believed that firms should pay a patent to local municipalities. A modification to the royalty proposal was introduced by the government, ensuring regional allocation of resources derived from the royalty. The law finally passed. But the government made several concessions. Among them, the tax would only apply to new investments, and they would have tax invariability for 15 years. Nonetheless, UDI legislators voted against the initiative, arguing a turning point on tax neutrality policies and deterioration of Chile’s international image. From the start, the initiative raised opposition. The main opponents were large mining firms grouped in Consejo Minero, the National Mining Society (SONAMI), whose members are mostly medium and small size firms, large business associations, like CPC and Sofofa, which claimed they opposed establishing a precedent. The think tank LyD also opposed the initiative on several grounds: 1) having a sector tax undermined the principle of tax neutrality; 2) setting this tax would induce a similar tax on other productive sectors; and 3) this initiative would impose a change in FDI contracts (under Decree Law 600), negatively affecting foreign investment. The royalty law failed when it only garnered 61 of the 69 votes needed in the Chamber of Deputies. After six months, the government sent a new draft law, this time with a specific tax to mining


exploitation rather than a tax on income so as not to modify agreements on concessions and to require only a simple majority in Congress for approval. Final Outcome: By 2005, conditions had changed (Boeninger, 2007). First, the royalty issue became more visible as public opinion overwhelmingly supported it. Second, taxing mining activities could be an election issue in 2005. In addition, the government was interested in solving this issue as a major confrontation on taxing mining might effectively affect FDI decisions and economic development as a whole. A window of opportunity led Concertación and RN legislators to agree on a solution. The opposition conceded to set 5 percent specific tax on income while the government agreed to a few concessions. For income tax purposes, the new tax would be an expenditure needed to produce the rent. The tax would be applied to operational rents, meaning sales costs, cumulative losses, and accelerated depreciation mechanisms would be considered in its calculation. No variation in tax regimes were offered for new investors for 15 years. There would be no retroactivity in the application for current contracts, although firms could choose between adopting the new tax regime and waiting until December of 2007 with a reduced tax of 4 percent. It may be argued that at the time of discussion, the visibility of the debate was very high. Public opinion was overwhelmingly in favor of a new tax on mining. The new tax also affected only a small but powerful group of firms, most of them foreign. Indeed, it was an operational rather than a political problem, since historic compromises through the FDI legal setup was in jeopardy. But once the royalty figure was changed to an ad valorem tax, there were no major ideological issues remaining at the time of voting. The approval of the royalty was the result of bilateral negotiations between the Ministry of Finance representing the government and opposition, supported by the mining companies and business associations. Both parties had political costs, given the electoral calendar and the time horizon. The government’s goal was to invite rather than force firms to switch to the new tax system. Meanwhile, the Alianza could not oppose a tax already in place in most mining countries that would provide resources for research and development and innovation. 6.5.2 Second Mixed Success Policy Reform: The Innovation Council Once the specific mining tax started to operate, questions arose as to how those resources would be allocated. One of the arguments advanced to justify the approval of the new tax was that it


would raise revenues to finance innovation. In addition, geographic considerations should also be in place when allocating the resources, with mining areas benefiting more than other areas. During 2005, when discussing the royalty, an indication to the law created an Innovation Fund (FIC) that would be funded with resources from the mining tax. In addition, the initiative also suggested the creation of a National Innovation Council that, among other things, would oversee the proper allocation of these resources. Since the Council would have an advisory role, it was also proposed that a more executive oriented government institution should be created to oversee the systemic consistency of the proposed instruments and programs. The initiative also included the creation of a Ministerial Committee for Innovation (CMI). Although articles related to the ad valorem tax were finally approved, those related to the creation of an Innovation National Council and to the Innovation Fund are still under debate. At least two issues related to these articles impede their approval. The first is the composition of the Council. Both sides agree that regional interests should be represented, although there is no clear view as to how this will be implemented. The other issue is that resources coming from the fund should be directly allocated at the regional level. Although there is an article suggesting that, there is no agreement about the percentage of the fund under this regime. The discussion is not between parties but among regions. It is not an ideological debate, but rather a geographical interest especially between mining regions and others. At the same time, since the Council is de facto in operation and the Fund is de facto created, there is no reason for the government to speed up the process. Although members of the Council are aware of its legal constitution, the idea is in jeopardy if not considered by the next administration. Main Players and Arguments: The initiative was proposed to Congress by the Ministry of Finance with support from the Ministry of Economics, since the latter should be the head of the CMI and therefore would be closely involved in the allocation and following of the FIC resources among different programs. Meanwhile, the National Council was created by a decree, mirroring what was already a legal initiative. At the same time, as part of the decree, the Council was asked to make suggestions on the use of the FIC resources already in place. In performing this duty, geographic considerations should be taken into account as if the law were already operating. That is, 35 percent of the FIC resources should be decided at the local rather than the national level.


Congress also has a role to play. On the one hand, legislators, especially those representing regional interests, push for a higher ratio of FIC resources to be locally assigned. On the other hand, they see the Council as an institutional competitor, since they were de facto deciding on how to allocate FIC resources. The Ministry of Finance, interested in the first part of the law, related to the financing, does not feel comfortable conceding special rights to local agencies in the allocation of public resources. Moreover, although the Council is a potential collaborator of Finance, since they are giving informed advice in FIC allocation, the creation of a State, rather than government institution, is not favored by Finance. Therefore, the best situation is having the council but with the alternative that a new government may change it if wanted with no major transaction costs. Current Situation: Today the initiative is still in Congress. For more than five years, it has been stuck at different levels of the congressional process. There are two elements that may explain this apparent win-win equilibrium. On the one hand, Finance does not like to release allocation power to local agencies, as it is not very confident about local capacity. On the other hand, the Council as an institution seems to be competing with legislators on their duties. So, none of the major players seems interested in discussing the initiative. The parliamentary side is composed of most of the legislators, since most of them have regional interests. Probably, the Council is the only agent interested in the legal initiative being delivered. Given that it has no permanent legal status and the documents explicitly mention that a permanent entity seeking a national strategy in innovation should be in place, non-government members are pushing Finance to accelerate the debate over this in Congress. Finally, the issue itself is not very visible to the public, since the problem is complicated and the solutions are far from clear. There is no clear answer to the institutional design challenge. Moreover, the Council has studied the subject and agreed that the institutional setup in promoting innovation needs to be tailor-made, depending among other things on several countryspecific characteristics. Therefore, it seems that the current equilibrium is quite stable and could only be broken if at some point the Council were dismantled. This might occur when the issue emerges again in the public arena and all actors are forced to make their positions known. Final Outcome: In 2005, in order to finance its health reform (Plan Auge) and social plan (Chile Solidario) and given the reduction in revenues because of the implementation of free trade agreements, the government was searching for new sources of revenue. Concertaci贸n parties did


not back a raise on the VAT, oil, or tobacco taxes as a way to finance these reforms. The opportunity arose when the press reported that large multinational firms (MNCs) were evading taxes. This became clear when the agreed price of a mine transacted among two large foreign MNCs failed to reflect economic losses reported to the Chilean tax office by the seller during previous ten years. Early in 2004, the Ministry of Finance sent legislation forward to modify the income tax law. It consisted of adding an ad valorem tax to sales which only applied to mining firms. Under the argument that firms were exploiting a non-renewable resource and not paying for it through the proposal, the government claimed that resources obtained from the tax would finance renewable activities (knowledge-based ones). The initiative led by Finance and backed by the Concertación was strongly opposed by the Alianza, which was pressured by the large mining firm association for obvious reasons. Arguments from the opposition included the danger of setting up sector-specific tax regimes and unilateral modification of MNC contracts with the state. These were mainly suggested by LyD, the think tank that articulated the Alianza arguments. However due to the fact that mainly foreign firms would be affected by the initiative, and given the public concern about the issue, the opposition soon agreed, as long as the law did not apply to past foreign mining contracts. Although the government was interested in financing these programs, there was still the issue of supporting knowledge-based activities, as explicitly argued by Finance. The law included the creation of an Innovation Fund (FIC) that would be financed with a portion of the resources collected by the tax and managed by an ad hoc council. It also suggested that a portion of the FIC’s resources be distributed geographically in accordance with the amount contributed by each region. This part of the law did not pass. After five years it is still under discussion in Congress. Meanwhile, by decree (a non-legislative ruling), the government created the National Council of Innovation, mimicking the institutional proposition defined in the draft law. Since that time, the Council has been de facto allocating resources from the FIC, taking into account among other things the regional partition also included in the law. The government not only directly participates in the Council board, since five ministers have voting rights, but also designates the other seven board members from civil society.


In this new equilibrium, Finance is no longer interested in pushing through the second part of the law. Although the existence of the Council may help in providing technical support and also controlling for the proper use of FIC resources in public agencies such as CORFO and CONICYT, pushing its legal constitution may expose Finance to a discussion of the allocation of FIC resources from regional congressmen, therefore losing control over its use. On the other hand, legislators see the Council as a competitor, given that is suggesting FIC allocations, including to regional programs. In this context, the only ones interested in passing the law are the non-government members of the Council. They see that the only way that programs and their related budgets can be sustained in the future is by the non-government Council, which can tackle the time inconsistency problems that governments and legislators face. Today, neither the Council nor the innovation issue is visible to the general public. This equilibrium is unlikely to change. However, the abolition of the Council may expose the situation especially if the incoming government is from the current opposition. Then, it will most likely continue operating under a presidential decree until another event occurs that shows that that this new equilibrium is not optimal.

6.Conclusion In the last 20 years, Chile’s policymaking performance has been outstanding in Latin America, transforming the country into an upper-middle-income economy. In the last few years, however, institutional problems have prevented the implementation of efficiency enhancing policy reforms much needed to maintain high productivity growth rates. This paper suggests that to regain high productivity growth rates, Chile needs to implement reforms to increase the transparency of its policymaking arenas and reduce spatial policy fragmentation, including the following: •

A far-reaching state reform geared to increase the efficiency of the state bureaucracy and its policy coordination.

An electoral system reform that maintains the positive effects of the binominal system while increasing it competitiveness.

Improvement of the legislation that regulates the internal workings of political parties, campaign finance, and political conflict of interests.


These conclusions present interesting lessons for other developing countries and for the economic development literature in general. They emphasize the importance of seemingly unrelated political and bureaucratic reforms for generating long-term sustainable growth. As this paper is being written, several major institutional reforms are under discussion in Congress, with strong possibilities for approval. If those reforms are implemented, Chile would become an interesting natural experiment to test the findings of this paper. These reforms would increase the visibility of moves of key socioeconomic actors, therefore increasing the likelihood of Chile regaining high productivity and economic growth. The PMP in Chile functioned well because of the high level of aggregation and the longterm horizons during the time the reforms were implemented did not affect strong and highly organized interests at the core of the two polarized extremes of the political spectrum. For different reasons, those polarized elites had evolved during the post-dictatorial period into accepting the principles of fiscal responsibility, rejecting populist economic policies, and embracing market-friendly economic policies (in part because the agrarian sector in Chile, unlike what happened in Brazil and Argentina, never lost its hegemonic power, and the labor movement focused on the public sector and non-tradable goods, not on import-substitution industries). A good check and balance system and the electoral system with its majoritarian incentives helped enforce political agreements to sustain fiscal responsibility and market-friendly policies. The electoral system and the opaqueness of the political and campaign finance legislation ended up protecting the political forces at both ends of the political spectrum, making it more difficult for real competition to emerge in the political arena. Pending reforms now would directly affect those status quo powers which, logically, block and oppose such reforms. Because political reforms have lagged behind, the challenge ahead is to deepen and expand democracy so that the interests of the unorganized majority can be effectively represented in the political arena. Political reforms must go in tandem with, if not ahead of, policy reforms.


References Aninat, C., J. Londregan, P. Navia and J. Vial. 2008. “Political Institutions, Policymaking Processes, and Policy Outcomes in Chile.” in E. Stein y M. Tommasi (eds.), Policymaking in Latin America: How Politics Shape Policy. Cambridge: David Rockefeller Center for Latin American Studies, Harvard University. Aninat, C. and J. Londregan. 2004. “Non-Pigouvian Payoffs and the Politics of Agricultural Trade Liberalization in Chile.” Presented at the Midwest Political Science Association 62nd Annual Meeting, April, Chicago. Boeninger, Edgardo. 1997. Democracia en Chile. Lecciones para la gobernabilidad. Santiago: Editorial Andrés Bello. Boeninger, Edgardo. 2007. Políticas públicas en democracia. Institucionalidad y experiencia chilena 1990-2006. Santiago: Uqbar Editores. Brunner, J. J., G. Elacqua, S. González, A. M. Montoya, and F. Salazar. 2006. Calidad de la educación. Claves para el debate. Santiago: RIL Editores. Cámara de Diputados. 1996. Legislatura Ordinaria 333, Sesión 26. Boletín de Sesiones. Congreso Nacional, Chile. Carey, J.M. 2002. “Parties, Coalitions, and the Chilean Congress in the 1990s” in S. Morgerstern and B. Nacif (eds.) Legislative Politics in Latin America. Boston: Cambridge University Press. Carmona, C. 2003. (Constitutional advisor to the Presidency, 1990-2008). Interview with C. Aninat, J. Londregan and J. Vial. Santiago, Chile. Comité Interministerial de Modernización de la Gestión Pública. 2000. El Estado al servicio de la gente. Balance 1994-2000. Secretaría General de la Presidencia, Santiago. Consejo Nacional de Innovación para la Competitividad. 2007 and 2008. “Hacia una Estrategia Nacional de Innovación para la Competitividad” Volume I and II. Cox, Cristian. 2004. "Las políticas educacionales de Chile en las últimas dos décadas del Siglo XX." In Políticas educacionales en el cambio de siglo. La reforma del sistema escolar de Chile, ed. C. Cox. Santiago: Editorial Universitaria. Dirección de Presupuesto (DIPRES). 2002. Informe de Finanzas Públicas. Proyecto de Ley de Presupuestos del Sector Público del año 2003. Presentado por Mario Marcel, Director de Presupuesto, October, Santiago, Chile.


----. 2003. Sistema de Control de Gestión y Presupuesto por Resultados. La Experiencia Chilena. Ministry of Finance, Santiago, Chile. Londregan, J. 2000. Legislative Institutions and Ideology in Chile. New York: Cambridge University Press. Ministerio de Obras Públicas, Transportes y Telecomunicaciones (MOPTT). 2003. Sistema de Concesiones en Chile: 1990-2003. Santiago, Chile. Murillo, María Victoria, Carlos Scartascini, and Mariano Tommasi. 2008. “The Political Economy of Productivity: Actors, Arenas, and Policies. A Framework of Analysis.” Research Department Working Paper 640. Washington, DC, United States: Inter American Development Bank. Servicio de Impuestos Internos. 2000. “La Administración Tributaria en la década de los noventa y el Proyecto de Ley de Lucha Contra la Evasión 2000-2005.” Departamento de Estudios, Santiago. Solow, Robert (1957): "Technical Change and the Aggregate Production Function." Review of Economics and Statistics, 39:312-320. Spiller, P. and M. Tommasi. 2003. “The Institutional Foundations of Public Policy: A Transactions Approach with Application to Argentina.” Journal of Law, Economics and Organization 19(2): 281-306. Spiller, P., E. Stein, and M. Tommasi. 2003. “Political Institutions, Policymaking Processes and Policy Outcomes: An Intertemporal Transactions Framework.” Design Paper #1 for the Red de Centros Project “Political Institutions, Policymaking Processes and Policy Outcomes.” Washington, D.C.: Inter-American Development Bank. Transparency International. 2003. Corruption Perceptions Index 2003. Berlin, Germany. ----. 2001. “Institucionalidad y Desempeño Fiscal: Una Mirada a la Experiencia de Chile en los 90.” Programa de Investigaciones de la Corporación de Estudios para Latinoamérica (CIEPLAN). Santiago, Chile. World Bank. 2004. Doing Business in 2004: Understanding Regulation. Washington, D.C.: The World Bank/Oxford University Press. World Economic Forum. 2003. The Global Competitiveness Report 2002-2003. New York: Oxford University Press.


World Economic Forum. 2008. The Global Competitiveness Report 2007-2008. New York: Oxford University Press. Zaldívar Larraín, Andrés. 1995. La transición inconclusa. Santiago: Editorial Los Andes. 


the political economy of productivity: the case of chile